public inbox for [email protected]help / color / mirror / Atom feed
[PATCH v24 3/5] Make archiver process an auxiliary process 279+ messages / 6 participants [nested] [flat]
* [PATCH v24 3/5] Make archiver process an auxiliary process @ 2018-11-07 07:53 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2018-11-07 07:53 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data wes moved onto shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/bootstrap/bootstrap.c | 8 +++ src/backend/postmaster/pgarch.c | 98 +++++++---------------------- src/backend/postmaster/pgstat.c | 6 ++ src/backend/postmaster/postmaster.c | 35 ++++++++--- src/include/miscadmin.h | 2 + src/include/pgstat.h | 1 + src/include/postmaster/pgarch.h | 4 +- 7 files changed, 67 insertions(+), 87 deletions(-) diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 8ea033610d..6e38f9a3d2 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -329,6 +329,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case BgWriterProcess: statmsg = pgstat_get_backend_desc(B_BG_WRITER); break; + case ArchiverProcess: + statmsg = pgstat_get_backend_desc(B_ARCHIVER); + break; case CheckpointerProcess: statmsg = pgstat_get_backend_desc(B_CHECKPOINTER); break; @@ -456,6 +459,11 @@ AuxiliaryProcessMain(int argc, char *argv[]) BackgroundWriterMain(); proc_exit(1); /* should never return */ + case ArchiverProcess: + /* don't set signals, archiver has its own agenda */ + PgArchiverMain(); + proc_exit(1); /* should never return */ + case CheckpointerProcess: /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index f84f882c4c..4342ebdab4 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -77,7 +77,6 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* @@ -96,7 +95,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); static void pgarch_exit(SIGNAL_ARGS); static void ArchSigHupHandler(SIGNAL_ARGS); static void ArchSigTermHandler(SIGNAL_ARGS); @@ -114,75 +112,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -222,8 +151,8 @@ pgarch_forkexec(void) * The argc/argv parameters are valid only in EXEC_BACKEND case. However, * since we don't use 'em, it hardly matters... */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -255,8 +184,27 @@ PgArchiverMain(int argc, char *argv[]) static void pgarch_exit(SIGNAL_ARGS) { - /* SIGQUIT means curl up and die ... */ - exit(1); + PG_SETMASK(&BlockSig); + + /* + * We DO NOT want to run proc_exit() callbacks -- we're here because + * shared memory may be corrupted, so we don't want to try to clean up our + * transaction. Just nail the windows shut and get out of town. Now that + * there's an atexit callback to prevent third-party code from breaking + * things by calling exit() directly, we have to reset the callbacks + * explicitly to make this work as intended. + */ + on_exit_reset(); + + /* + * Note we do exit(2) not exit(0). This is to force the postmaster into a + * system reset cycle if some idiot DBA sends a manual SIGQUIT to a random + * backend. This is necessary precisely because we don't clean up our + * shared memory state. (The "dead man switch" mechanism in pmsignal.c + * should ensure the postmaster sees this as a crash, too, but no harm in + * being doubly sure.) + */ + exit(2); } /* SIGHUP signal handler for archiver process */ diff --git a/src/backend/postmaster/pgstat.c b/src/backend/postmaster/pgstat.c index fabcf31de8..8299d2a435 100644 --- a/src/backend/postmaster/pgstat.c +++ b/src/backend/postmaster/pgstat.c @@ -2932,6 +2932,9 @@ pgstat_bestart(void) case StartupProcess: lbeentry.st_backendType = B_STARTUP; break; + case ArchiverProcess: + beentry->st_backendType = B_ARCHIVER; + break; case BgWriterProcess: lbeentry.st_backendType = B_BG_WRITER; break; @@ -4275,6 +4278,9 @@ pgstat_get_backend_desc(BackendType backendType) switch (backendType) { + case B_ARCHIVER: + backendDesc = "archiver"; + break; case B_AUTOVAC_LAUNCHER: backendDesc = "autovacuum launcher"; break; diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 9ff2832c00..84fda38249 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -146,7 +146,8 @@ #define BACKEND_TYPE_AUTOVAC 0x0002 /* autovacuum worker process */ #define BACKEND_TYPE_WALSND 0x0004 /* walsender process */ #define BACKEND_TYPE_BGWORKER 0x0008 /* bgworker process */ -#define BACKEND_TYPE_ALL 0x000F /* OR of all the above */ +#define BACKEND_TYPE_ARCHIVER 0x0010 /* archiver process */ +#define BACKEND_TYPE_ALL 0x001F /* OR of all the above */ #define BACKEND_TYPE_WORKER (BACKEND_TYPE_AUTOVAC | BACKEND_TYPE_BGWORKER) @@ -539,6 +540,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1787,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3042,7 +3044,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3187,10 +3189,8 @@ reaper(SIGNAL_ARGS) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3438,7 +3438,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3643,6 +3643,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3915,6 +3927,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5190,7 +5203,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5475,6 +5488,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index bc6e03fbc7..1f4db67f3f 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -399,6 +399,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -411,6 +412,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/pgstat.h b/src/include/pgstat.h index fe076d823d..65713abc2b 100644 --- a/src/include/pgstat.h +++ b/src/include/pgstat.h @@ -718,6 +718,7 @@ typedef struct PgStat_GlobalStats */ typedef enum BackendType { + B_ARCHIVER, B_AUTOVAC_LAUNCHER, B_AUTOVAC_WORKER, B_BACKEND, diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index 2474eac26a..88f16863d4 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ -- 2.23.0 ----Next_Part(Tue_Dec__3_17_27_59_2019_887)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v24-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH 3/5] Make archiver process an auxiliary process @ 2018-11-07 07:53 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2018-11-07 07:53 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data wes moved onto shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/bootstrap/bootstrap.c | 8 +++ src/backend/postmaster/pgarch.c | 98 +++++++---------------------- src/backend/postmaster/pgstat.c | 6 ++ src/backend/postmaster/postmaster.c | 43 ++++++++----- src/include/miscadmin.h | 2 + src/include/pgstat.h | 1 + src/include/postmaster/pgarch.h | 4 +- 7 files changed, 70 insertions(+), 92 deletions(-) diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index bfc629c753..2f38420313 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -326,6 +326,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: statmsg = pgstat_get_backend_desc(B_STARTUP); break; + case ArchiverProcess: + statmsg = pgstat_get_backend_desc(B_ARCHIVER); + break; case BgWriterProcess: statmsg = pgstat_get_backend_desc(B_BG_WRITER); break; @@ -451,6 +454,11 @@ AuxiliaryProcessMain(int argc, char *argv[]) StartupProcessMain(); proc_exit(1); /* should never return */ + case ArchiverProcess: + /* don't set signals, archiver has its own agenda */ + PgArchiverMain(); + proc_exit(1); /* should never return */ + case BgWriterProcess: /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 3ca30badb2..6441e69e9a 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -78,7 +78,6 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* @@ -95,7 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); static void pgarch_exit(SIGNAL_ARGS); static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); @@ -111,75 +109,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -219,8 +148,8 @@ pgarch_forkexec(void) * The argc/argv parameters are valid only in EXEC_BACKEND case. However, * since we don't use 'em, it hardly matters... */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -252,8 +181,27 @@ PgArchiverMain(int argc, char *argv[]) static void pgarch_exit(SIGNAL_ARGS) { - /* SIGQUIT means curl up and die ... */ - exit(1); + PG_SETMASK(&BlockSig); + + /* + * We DO NOT want to run proc_exit() callbacks -- we're here because + * shared memory may be corrupted, so we don't want to try to clean up our + * transaction. Just nail the windows shut and get out of town. Now that + * there's an atexit callback to prevent third-party code from breaking + * things by calling exit() directly, we have to reset the callbacks + * explicitly to make this work as intended. + */ + on_exit_reset(); + + /* + * Note we do exit(2) not exit(0). This is to force the postmaster into a + * system reset cycle if some idiot DBA sends a manual SIGQUIT to a random + * process. This is necessary precisely because we don't clean up our + * shared memory state. (The "dead man switch" mechanism in pmsignal.c + * should ensure the postmaster sees this as a crash, too, but no harm in + * being doubly sure.) + */ + exit(2); } /* SIGUSR1 signal handler for archiver process */ diff --git a/src/backend/postmaster/pgstat.c b/src/backend/postmaster/pgstat.c index 51c486bebd..ca5c6376e5 100644 --- a/src/backend/postmaster/pgstat.c +++ b/src/backend/postmaster/pgstat.c @@ -2927,6 +2927,9 @@ pgstat_bestart(void) case StartupProcess: lbeentry.st_backendType = B_STARTUP; break; + case ArchiverProcess: + beentry->st_backendType = B_ARCHIVER; + break; case BgWriterProcess: lbeentry.st_backendType = B_BG_WRITER; break; @@ -4273,6 +4276,9 @@ pgstat_get_backend_desc(BackendType backendType) switch (backendType) { + case B_ARCHIVER: + backendDesc = "archiver"; + break; case B_AUTOVAC_LAUNCHER: backendDesc = "autovacuum launcher"; break; diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7a92dac525..ee41d7009e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -146,7 +146,8 @@ #define BACKEND_TYPE_AUTOVAC 0x0002 /* autovacuum worker process */ #define BACKEND_TYPE_WALSND 0x0004 /* walsender process */ #define BACKEND_TYPE_BGWORKER 0x0008 /* bgworker process */ -#define BACKEND_TYPE_ALL 0x000F /* OR of all the above */ +#define BACKEND_TYPE_ARCHIVER 0x0010 /* archiver process */ +#define BACKEND_TYPE_ALL 0x001F /* OR of all the above */ #define BACKEND_TYPE_WORKER (BACKEND_TYPE_AUTOVAC | BACKEND_TYPE_BGWORKER) @@ -539,6 +540,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1787,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3042,7 +3044,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3177,20 +3179,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3438,7 +3436,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3643,6 +3641,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3915,6 +3925,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5190,7 +5201,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5475,6 +5486,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 62d64aa0a1..dc7dd1c164 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -397,6 +397,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -409,6 +410,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/pgstat.h b/src/include/pgstat.h index 36b530bc27..0334213b98 100644 --- a/src/include/pgstat.h +++ b/src/include/pgstat.h @@ -718,6 +718,7 @@ typedef struct PgStat_GlobalStats */ typedef enum BackendType { + B_ARCHIVER, B_AUTOVAC_LAUNCHER, B_AUTOVAC_WORKER, B_BACKEND, diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ -- 2.23.0 ----Next_Part(Wed_Jan_22_17_24_04_2020_620)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v33 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cdd586fcfb..ee3444284b 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..063d1323ea 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 160afe9f39..9de9396628 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -539,6 +539,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1786,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3068,7 +3069,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3203,20 +3204,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3464,7 +3461,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3669,6 +3666,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3941,6 +3950,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5216,7 +5226,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5259,16 +5269,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5501,6 +5501,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 5aa19d3f78..7acc48734e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -187,6 +187,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index e917dfe92d..0783692c83 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -336,6 +336,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index ae4f573ab4..1a8a0c2e15 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -272,6 +272,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.2 ----Next_Part(Fri_May_15_17_30_36_2020_111)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v33-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v29 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlog.c | 46 ++++++++++ src/backend/access/transam/xlogarchive.c | 2 +- src/backend/bootstrap/bootstrap.c | 22 ++--- src/backend/postmaster/pgarch.c | 102 ++++------------------- src/backend/postmaster/postmaster.c | 53 ++++++------ src/include/access/xlog.h | 2 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - 10 files changed, 108 insertions(+), 127 deletions(-) diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c index 977d448f50..d75a257da2 100644 --- a/src/backend/access/transam/xlog.c +++ b/src/backend/access/transam/xlog.c @@ -684,6 +684,13 @@ typedef struct XLogCtlData */ Latch recoveryWakeupLatch; + /* + * archiverWakeupLatch is used to wake up the archiver process to process + * completed WAL segments, if it is waiting for WAL to arrive. Protected + * by info_lck. + */ + Latch *archiverWakeupLatch; + /* * During recovery, we keep a copy of the latest checkpoint record here. * lastCheckPointRecPtr points to start of checkpoint record and @@ -8450,6 +8457,45 @@ GetLastSegSwitchData(XLogRecPtr *lastSwitchLSN) return result; } +/* + * XLogArchiveWakeupStart - Set up archiver wakeup stuff + */ +void +XLogArchiveWakeupStart(void) +{ + SpinLockAcquire(&XLogCtl->info_lck); + Assert(XLogCtl->archiverWakeupLatch == NULL); + XLogCtl->archiverWakeupLatch = MyLatch; + SpinLockRelease(&XLogCtl->info_lck); +} + +/* + * XLogArchiveWakeupEnd - Clean up archiver wakeup stuff + */ +void +XLogArchiveWakeupEnd(void) +{ + SpinLockAcquire(&XLogCtl->info_lck); + XLogCtl->archiverWakeupLatch = NULL; + SpinLockRelease(&XLogCtl->info_lck); +} + +/* + * XLogWakeupArchiver - Wake up archiver process + */ +void +XLogArchiveWakeup(void) +{ + Latch *latch; + + SpinLockAcquire(&XLogCtl->info_lck); + latch = XLogCtl->archiverWakeupLatch; + SpinLockRelease(&XLogCtl->info_lck); + + if (latch) + SetLatch(latch); +} + /* * This must be called ONCE during postmaster or standalone-backend shutdown */ diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index d62c12310a..d6c8bcce8b 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -490,7 +490,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + XLogArchiveWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..6fe7a136ba 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,7 +79,6 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* @@ -95,8 +95,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +108,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +141,21 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ +/* Clean up notification stuff on exit */ +static void +PgArchiverKill(int code, Datum arg) +{ + XLogArchiveWakeupEnd(); +} + /* * PgArchiverMain * * The argc/argv parameters are valid only in EXEC_BACKEND case. However, * since we don't use 'em, it hardly matters... */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,7 +167,7 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); @@ -240,24 +176,14 @@ PgArchiverMain(int argc, char *argv[]) MyBackendType = B_ARCHIVER; init_ps_display(NULL); + XLogArchiveWakeupStart(); + on_shmem_exit(PgArchiverKill, 0); + pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 73d278f3b2..a4b9d212a2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -146,7 +146,8 @@ #define BACKEND_TYPE_AUTOVAC 0x0002 /* autovacuum worker process */ #define BACKEND_TYPE_WALSND 0x0004 /* walsender process */ #define BACKEND_TYPE_BGWORKER 0x0008 /* bgworker process */ -#define BACKEND_TYPE_ALL 0x000F /* OR of all the above */ +#define BACKEND_TYPE_ARCHIVER 0x0010 /* archiver process */ +#define BACKEND_TYPE_ALL 0x001F /* OR of all the above */ #define BACKEND_TYPE_WORKER (BACKEND_TYPE_AUTOVAC | BACKEND_TYPE_BGWORKER) @@ -539,6 +540,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1787,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3055,7 +3057,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3190,20 +3192,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3451,7 +3449,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3656,6 +3654,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3928,6 +3938,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5208,7 +5219,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5251,16 +5262,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5493,6 +5494,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 9ec7b31cce..c0e149d526 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -313,6 +313,8 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ -- 2.18.2 ----Next_Part(Wed_Apr__1_15_15_11_2020_923)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v29-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v30 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlog.c | 46 +++++++++ src/backend/access/transam/xlogarchive.c | 2 +- src/backend/bootstrap/bootstrap.c | 22 +++-- src/backend/postmaster/pgarch.c | 115 ++++------------------- src/backend/postmaster/postmaster.c | 53 ++++++----- src/include/access/xlog.h | 2 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - 10 files changed, 112 insertions(+), 136 deletions(-) diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c index 977d448f50..d75a257da2 100644 --- a/src/backend/access/transam/xlog.c +++ b/src/backend/access/transam/xlog.c @@ -684,6 +684,13 @@ typedef struct XLogCtlData */ Latch recoveryWakeupLatch; + /* + * archiverWakeupLatch is used to wake up the archiver process to process + * completed WAL segments, if it is waiting for WAL to arrive. Protected + * by info_lck. + */ + Latch *archiverWakeupLatch; + /* * During recovery, we keep a copy of the latest checkpoint record here. * lastCheckPointRecPtr points to start of checkpoint record and @@ -8450,6 +8457,45 @@ GetLastSegSwitchData(XLogRecPtr *lastSwitchLSN) return result; } +/* + * XLogArchiveWakeupStart - Set up archiver wakeup stuff + */ +void +XLogArchiveWakeupStart(void) +{ + SpinLockAcquire(&XLogCtl->info_lck); + Assert(XLogCtl->archiverWakeupLatch == NULL); + XLogCtl->archiverWakeupLatch = MyLatch; + SpinLockRelease(&XLogCtl->info_lck); +} + +/* + * XLogArchiveWakeupEnd - Clean up archiver wakeup stuff + */ +void +XLogArchiveWakeupEnd(void) +{ + SpinLockAcquire(&XLogCtl->info_lck); + XLogCtl->archiverWakeupLatch = NULL; + SpinLockRelease(&XLogCtl->info_lck); +} + +/* + * XLogWakeupArchiver - Wake up archiver process + */ +void +XLogArchiveWakeup(void) +{ + Latch *latch; + + SpinLockAcquire(&XLogCtl->info_lck); + latch = XLogCtl->archiverWakeupLatch; + SpinLockRelease(&XLogCtl->info_lck); + + if (latch) + SetLatch(latch); +} + /* * This must be called ONCE during postmaster or standalone-backend shutdown */ diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index d62c12310a..d6c8bcce8b 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -490,7 +490,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + XLogArchiveWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..7b66cb5613 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,7 +79,6 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* @@ -95,8 +95,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +108,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +141,16 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Clean up notification stuff on exit */ +static void +PgArchiverKill(int code, Datum arg) +{ + XLogArchiveWakeupEnd(); +} + +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +162,23 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); - + XLogArchiveWakeupStart(); + on_shmem_exit(PgArchiverKill, 0); + pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 73d278f3b2..a4b9d212a2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -146,7 +146,8 @@ #define BACKEND_TYPE_AUTOVAC 0x0002 /* autovacuum worker process */ #define BACKEND_TYPE_WALSND 0x0004 /* walsender process */ #define BACKEND_TYPE_BGWORKER 0x0008 /* bgworker process */ -#define BACKEND_TYPE_ALL 0x000F /* OR of all the above */ +#define BACKEND_TYPE_ARCHIVER 0x0010 /* archiver process */ +#define BACKEND_TYPE_ALL 0x001F /* OR of all the above */ #define BACKEND_TYPE_WORKER (BACKEND_TYPE_AUTOVAC | BACKEND_TYPE_BGWORKER) @@ -539,6 +540,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1787,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3055,7 +3057,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3190,20 +3192,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3451,7 +3449,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3656,6 +3654,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3928,6 +3938,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5208,7 +5219,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5251,16 +5262,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5493,6 +5494,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 9ec7b31cce..c0e149d526 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -313,6 +313,8 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ -- 2.18.2 ----Next_Part(Wed_Apr__1_17_37_23_2020_570)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v30-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v31 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index d62c12310a..2f8672ac0c 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..063d1323ea 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 73d278f3b2..e97995f973 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -539,6 +539,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1786,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3055,7 +3056,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3190,20 +3191,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3451,7 +3448,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3656,6 +3653,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3928,6 +3937,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5208,7 +5218,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5251,16 +5261,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5493,6 +5493,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 9938cddb57..af4599bd82 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -187,6 +187,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 9ec7b31cce..4435df82b6 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -313,6 +313,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index d21780108b..a87e7dc711 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -272,6 +272,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.2 ----Next_Part(Fri_Apr__3_17_31_17_2020_104)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v31-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v32 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index d62c12310a..2f8672ac0c 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..063d1323ea 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 73d278f3b2..e97995f973 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -539,6 +539,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1786,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3055,7 +3056,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3190,20 +3191,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3451,7 +3448,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3656,6 +3653,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3928,6 +3937,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5208,7 +5218,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5251,16 +5261,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5493,6 +5493,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 9938cddb57..af4599bd82 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -187,6 +187,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 9ec7b31cce..4435df82b6 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -313,6 +313,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index d21780108b..a87e7dc711 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -272,6 +272,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.2 ----Next_Part(Tue_Apr__7_16_38_17_2020_299)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v32-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v33 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 55becd65d4..2417de72a0 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..063d1323ea 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 0578e92ba9..2687d2af47 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -539,6 +539,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1786,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3055,7 +3056,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3190,20 +3191,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3451,7 +3448,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3656,6 +3653,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3928,6 +3937,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5203,7 +5213,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5246,16 +5256,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5488,6 +5488,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 5aa19d3f78..7acc48734e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -187,6 +187,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 0a12afb59e..099da8cd76 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -336,6 +336,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index ae4f573ab4..1a8a0c2e15 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -272,6 +272,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.2 ----Next_Part(Thu_Apr_30_13_22_05_2020_409)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v33-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v25 4/8] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/bootstrap/bootstrap.c | 22 +++++---- src/backend/postmaster/pgarch.c | 75 +---------------------------- src/backend/postmaster/postmaster.c | 43 +++++++++++------ src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- 5 files changed, 46 insertions(+), 100 deletions(-) diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..4971b3ae42 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -78,7 +78,6 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* @@ -95,7 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); @@ -110,75 +108,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -218,8 +147,8 @@ pgarch_forkexec(void) * The argc/argv parameters are valid only in EXEC_BACKEND case. However, * since we don't use 'em, it hardly matters... */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 2b9ab32293..cab7fb5381 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -146,7 +146,8 @@ #define BACKEND_TYPE_AUTOVAC 0x0002 /* autovacuum worker process */ #define BACKEND_TYPE_WALSND 0x0004 /* walsender process */ #define BACKEND_TYPE_BGWORKER 0x0008 /* bgworker process */ -#define BACKEND_TYPE_ALL 0x000F /* OR of all the above */ +#define BACKEND_TYPE_ARCHIVER 0x0010 /* archiver process */ +#define BACKEND_TYPE_ALL 0x001F /* OR of all the above */ #define BACKEND_TYPE_WORKER (BACKEND_TYPE_AUTOVAC | BACKEND_TYPE_BGWORKER) @@ -539,6 +540,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1787,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3055,7 +3057,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3190,20 +3192,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3451,7 +3449,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3656,6 +3654,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3928,6 +3938,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5208,7 +5219,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5493,6 +5504,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ -- 2.18.2 ----Next_Part(Thu_Mar_19_20_30_04_2020_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v25-0005-Use-latch-instead-of-SIGUSR1-to-wake-up-archiver.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v43 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index f39dc4ddf1..d69e20961d 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index a7ed93fdc1..edda899554 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 5d09822c81..8a4706a870 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -547,6 +547,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1787,7 +1788,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3041,7 +3042,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3176,20 +3177,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3437,7 +3434,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3642,6 +3639,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3938,6 +3947,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5185,7 +5195,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5230,16 +5240,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5481,6 +5481,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 7dc3911590..fc23539137 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index e77f76ae8a..a656910d02 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Dec_11_16_50_03_2020_915)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v43-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v44 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index f39dc4ddf1..d69e20961d 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index a7ed93fdc1..edda899554 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 5d09822c81..8a4706a870 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -547,6 +547,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1787,7 +1788,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3041,7 +3042,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3176,20 +3177,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3437,7 +3434,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3642,6 +3639,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3938,6 +3947,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5185,7 +5195,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5230,16 +5240,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5481,6 +5481,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 7dc3911590..fc23539137 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index e77f76ae8a..a656910d02 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Mon_Dec_21_17_16_20_2020_608)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v44-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v36 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 8f8734dc1d..2e6c322142 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 76b2f5066f..81bfaea869 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..063d1323ea 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 42223c0f61..4c296dd214 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -554,6 +554,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1799,7 +1800,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3051,7 +3052,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3186,20 +3187,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3447,7 +3444,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3652,6 +3649,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3948,6 +3957,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5223,7 +5233,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5268,16 +5278,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5515,6 +5515,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index aa9fbd8054..9cc5226667 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 9c9a50ae45..de20520b8c 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -345,6 +345,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.4 ----Next_Part(Tue_Sep__8_17_55_57_2020_197)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v36-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v37 3/6] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 8f8734dc1d..2e6c322142 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 76b2f5066f..81bfaea869 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 959e3b8873..b811c961a6 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -555,6 +555,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1800,7 +1801,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3054,7 +3055,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3189,20 +3190,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3450,7 +3447,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3655,6 +3652,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3951,6 +3960,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5230,7 +5240,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5275,16 +5285,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5526,6 +5526,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 88566bd9fa..746bed773e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 9c9a50ae45..de20520b8c 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -345,6 +345,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.4 ----Next_Part(Fri_Sep_25_09_27_26_2020_419)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v37-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v38 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cae93ab69d..6908bec2f9 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 76b2f5066f..81bfaea869 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 959e3b8873..b811c961a6 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -555,6 +555,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1800,7 +1801,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3054,7 +3055,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3189,20 +3190,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3450,7 +3447,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3655,6 +3652,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3951,6 +3960,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5230,7 +5240,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5275,16 +5285,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5526,6 +5526,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 88566bd9fa..746bed773e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 9c9a50ae45..de20520b8c 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -345,6 +345,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.4 ----Next_Part(Thu_Oct__1_09_07_22_2020_252)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v38-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v39 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cae93ab69d..6908bec2f9 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 76b2f5066f..81bfaea869 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 959e3b8873..b811c961a6 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -555,6 +555,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1800,7 +1801,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3054,7 +3055,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3189,20 +3190,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3450,7 +3447,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3655,6 +3652,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3951,6 +3960,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5230,7 +5240,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5275,16 +5285,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5526,6 +5526,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 88566bd9fa..746bed773e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 9c9a50ae45..de20520b8c 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -345,6 +345,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.4 ----Next_Part(Tue_Oct__6_10_06_44_2020_382)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v39-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v40 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cae93ab69d..6908bec2f9 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index a7ed93fdc1..edda899554 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 959e3b8873..b811c961a6 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -555,6 +555,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1800,7 +1801,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3054,7 +3055,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3189,20 +3190,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3450,7 +3447,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3655,6 +3652,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3951,6 +3960,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5230,7 +5240,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5275,16 +5285,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5526,6 +5526,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 88566bd9fa..746bed773e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 9c9a50ae45..de20520b8c 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -345,6 +345,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.4 ----Next_Part(Wed_Nov__4_17_39_10_2020_208)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v40-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v41 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cae93ab69d..6908bec2f9 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index a7ed93fdc1..edda899554 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 959e3b8873..b811c961a6 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -555,6 +555,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1800,7 +1801,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3054,7 +3055,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3189,20 +3190,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3450,7 +3447,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3655,6 +3652,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3951,6 +3960,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5230,7 +5240,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5275,16 +5285,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5526,6 +5526,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 88566bd9fa..746bed773e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 9c9a50ae45..de20520b8c 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -345,6 +345,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.4 ----Next_Part(Fri_Nov__6_09_27_56_2020_738)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v41-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v42 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cae93ab69d..6908bec2f9 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index a7ed93fdc1..edda899554 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index ed1b65358d..e3a520def9 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index b7799ed1d2..a1dd6964e9 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -547,6 +547,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1786,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3039,7 +3040,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3174,20 +3175,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3435,7 +3432,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3640,6 +3637,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3936,6 +3945,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5177,7 +5187,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5222,16 +5232,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5473,6 +5473,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 88566bd9fa..746bed773e 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -181,6 +181,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 221af87e71..65443063e8 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 72e3352398..bba002ad24 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -418,6 +418,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -430,6 +431,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 9c9a50ae45..de20520b8c 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -345,6 +345,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.4 ----Next_Part(Wed_Nov_11_10_07_22_2020_796)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v42-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v34 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cdd586fcfb..ee3444284b 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..063d1323ea 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 160afe9f39..9de9396628 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -539,6 +539,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1786,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3068,7 +3069,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3203,20 +3204,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3464,7 +3461,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3669,6 +3666,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3941,6 +3950,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5216,7 +5226,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5259,16 +5269,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5501,6 +5501,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index f5eef6fa4e..46f0198510 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -187,6 +187,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index e917dfe92d..0783692c83 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -336,6 +336,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 14fa127ab1..619b2f9c71 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 1ee9000b2b..8ab6d859bb 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -270,6 +270,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.2 ----Next_Part(Mon_Jun__1_18_00_01_2020_089)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v34-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v35 4/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index cdd586fcfb..ee3444284b 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -489,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 5480a024e0..d398ce6f03 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -319,6 +319,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -439,30 +442,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index 37be0e2bbb..063d1323ea 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) pqsignal(SIGQUIT, SignalHandlerForCrashExit); pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index b4d475bb0b..30f3fc6ba1 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -539,6 +539,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1785,7 +1786,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3068,7 +3069,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3203,20 +3204,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3464,7 +3461,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3669,6 +3666,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3941,6 +3950,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5216,7 +5226,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5259,16 +5269,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5501,6 +5501,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index f5eef6fa4e..46f0198510 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -187,6 +187,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index e917dfe92d..0783692c83 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -336,6 +336,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 1c67de2ede..54ce0b97d7 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 18bc8a7b90..cf0b463eba 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index b3200874ca..e3ffc63f14 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index 56c5ec4481..c691acf8cd 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 1ee9000b2b..8ab6d859bb 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -270,6 +270,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.18.2 ----Next_Part(Mon_Jun__8_17_32_04_2020_489)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v35-0005-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v45 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index db0cfaa360..aabf9d73eb 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Jan__8_10_24_34_2021_185)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v45-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v46 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 0786fcf103..430d438303 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -354,6 +354,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_14_15_14_25_2021_903)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v46-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v47 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index 7de27ee4e0..af91c313e2 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1788,7 +1789,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3027,7 +3028,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3162,20 +3163,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3423,7 +3420,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3629,6 +3626,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3932,6 +3941,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5160,7 +5170,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5214,16 +5224,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5465,6 +5465,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index c87ffc6549..a1e51c5b99 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index 683ab64f76..6cdaf3753d 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -355,6 +355,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Thu_Jan_21_12_03_48_2021_284)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v47-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v48 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++-- src/backend/postmaster/pgarch.c | 130 +++-------------------- src/backend/postmaster/postmaster.c | 50 +++++---- src/backend/storage/lmgr/proc.c | 1 + src/include/access/xlog.h | 3 + src/include/access/xlogarchive.h | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 4 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 3 + 11 files changed, 69 insertions(+), 154 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..d01859bde5 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,9 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" +#include "storage/latch.h" #include "storage/pmsignal.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +492,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..9a2e21bf86 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,13 +79,11 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- @@ -95,8 +94,6 @@ static volatile sig_atomic_t ready_to_stop = false; static pid_t pgarch_forkexec(void); #endif -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -110,75 +107,6 @@ static void pgarch_archiveDone(char *xlog); * ------------------------------------------------------------ */ -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - /* ------------------------------------------------------------ * Local functions called by archiver follow * ------------------------------------------------------------ @@ -212,14 +140,9 @@ pgarch_forkexec(void) #endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +154,26 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; -} - /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +198,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +236,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +258,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..22e1cd244c 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,16 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. + * Was it the archiver? Normal exit can be ignored; we'll start a new + * one at the next iteration of the postmaster's main loop, if + * necessary. Any other exit condition is treated as a crash. */ if (pid == PgArchPID) { PgArchPID = 0; if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + HandleChildCrash(pid, exitstatus, + _("archiver process")); continue; } @@ -3403,7 +3400,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,6 +3606,18 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) + { + ereport(DEBUG2, + (errmsg_internal("sending %s to process %d", + (SendStop ? "SIGSTOP" : "SIGQUIT"), + (int) PgArchPID))); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); + } + /* * Force a power-cycle of the pgarch process too. (This isn't absolutely * necessary, but it seems like a good idea for robustness, and it @@ -3912,6 +3921,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5140,7 +5150,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5204,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5445,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index 75ec1073bd..551f518cc2 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -339,6 +339,9 @@ extern XLogRecPtr GetRedoRecPtr(void); extern XLogRecPtr GetInsertRecPtr(void); extern XLogRecPtr GetFlushRecPtr(void); extern XLogRecPtr GetLastImportantRecPtr(void); +extern void XLogArchiveWakeupStart(void); +extern void XLogArchiveWakeupEnd(void); +extern void XLogArchiveWakeup(void); extern void RemovePromoteSignalFiles(void); extern bool PromoteIsTriggered(void); diff --git a/src/include/access/xlogarchive.h b/src/include/access/xlogarchive.h index 3edd1a976c..1a59181cf9 100644 --- a/src/include/access/xlogarchive.h +++ b/src/include/access/xlogarchive.h @@ -25,6 +25,7 @@ extern void ExecuteRecoveryCommand(const char *command, const char *commandName, extern void KeepFileRestoredFromArchive(const char *path, const char *xlogfname); extern void XLogArchiveNotify(const char *xlog); extern void XLogArchiveNotifySeg(XLogSegNo segno); +extern void XLogArchiveWakeup(void); extern void XLogArchiveForceDone(const char *xlog); extern bool XLogArchiveCheckDone(const char *xlog); extern bool XLogArchiveIsBusy(const char *xlog); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..385b002dfe 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -32,8 +32,6 @@ */ extern int pgarch_start(void); -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..491747c113 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,9 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; + /* Current shared estimate of appropriate spins_per_delay value */ } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; -- 2.27.0 ----Next_Part(Fri_Mar__5_17_18_56_2021_497)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v48-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v49 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_16_53_11_2021_575)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v49-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v50 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 174 +++-------------------- src/backend/postmaster/postmaster.c | 88 ++++++------ src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 83 insertions(+), 232 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..e237cedaff 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -48,6 +48,7 @@ #include "storage/latch.h" #include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,25 +79,17 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); @@ -105,121 +98,9 @@ static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +112,24 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..d9f8d82650 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3126,6 +3127,18 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? We treat it the same way to WAL receiver. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + continue; + } + /* * Was it the autovacuum launcher? Normal exit can be ignored; we'll * start a new one at the next iteration of the postmaster's main @@ -3141,24 +3154,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3403,7 +3398,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3604,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3904,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -4989,6 +4982,19 @@ SubPostmasterMain(int argc, char *argv[]) AuxiliaryProcessMain(argc - 2, argv + 2); /* does not return */ } + if (strcmp(argv[1], "--forkarch") == 0) + { + /* Restore basic shared memory pointers */ + InitShmemAccess(UsedShmemSegAddr); + + /* Need a PGPROC to run CreateSharedMemoryAndSemaphores */ + InitAuxiliaryProcess(); + + /* Attach process to shared data structures */ + CreateSharedMemoryAndSemaphores(); + + PgArchiverMain(); /* does not return */ + } if (strcmp(argv[1], "--forkavlauncher") == 0) { /* Restore basic shared memory pointers */ @@ -5037,12 +5043,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5140,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5194,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5435,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d73890b228 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -357,6 +357,8 @@ typedef struct PROC_HDR int startupProcPid; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Archiver process's latch */ + Latch *archiverLatch; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Tue_Mar__9_18_29_34_2021_806)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v50-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v51 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 7 +- src/backend/bootstrap/bootstrap.c | 22 +-- src/backend/postmaster/pgarch.c | 194 ++++------------------- src/backend/postmaster/postmaster.c | 83 +++++----- src/backend/storage/lmgr/proc.c | 1 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 10 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 10 +- 9 files changed, 94 insertions(+), 236 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..8b66a30931 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -29,7 +29,8 @@ #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -490,8 +491,8 @@ XLogArchiveNotify(const char *xlog) } /* Notify archiver that it's got something to do */ - if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + if (IsUnderPostmaster && ProcGlobal->archiverLatch) + SetLatch(ProcGlobal->archiverLatch); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c219780c60 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,13 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -78,148 +75,29 @@ * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; - - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); -} -#endif /* EXEC_BACKEND */ - - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,31 +109,27 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + ProcGlobal->archiverLatch = &MyProc->procLatch; pgarch_MainLoop(); - exit(0); -} - -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) -{ - int save_errno = errno; - - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + proc_exit(0); } /* SIGUSR2 signal handler for archiver process */ @@ -282,14 +156,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +194,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +216,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +602,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + ProcGlobal->archiverLatch = NULL; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..fd509869aa 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving of remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 897045ee27..bf5d6da89b 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -182,6 +182,7 @@ InitProcGlobal(void) ProcGlobal->startupBufferPinWaitBufId = -1; ProcGlobal->walwriterLatch = NULL; ProcGlobal->checkpointerLatch = NULL; + ProcGlobal->archiverLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..737ed69e57 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,6 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern void PgArchiverMain(void) pg_attribute_noreturn(); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..d71700a565 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -350,6 +350,8 @@ typedef struct PROC_HDR Latch *walwriterLatch; /* Checkpointer process's latch */ Latch *checkpointerLatch; + /* Archiver process's latch */ + Latch *archiverLatch; /* Current shared estimate of appropriate spins_per_delay value */ int spins_per_delay; /* The proc of the Startup process, since not in ProcArray */ @@ -370,11 +372,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_12_10_39_2021_432)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v51-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v52 3/7] Make archiver process an auxiliary process @ 2020-03-13 07:59 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2020-03-13 07:59 UTC (permalink / raw) This is a preliminary patch for shared-memory based stats collector. Archiver process must be a auxiliary process since it uses shared memory after stats data was moved into shared-memory. Make the process an auxiliary process in order to make it work. --- src/backend/access/transam/xlogarchive.c | 6 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 224 ++++++++--------------- src/backend/postmaster/postmaster.c | 83 ++++----- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- 9 files changed, 139 insertions(+), 222 deletions(-) diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..2558bcfb08 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,13 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" +#include "storage/latch.h" +#include "storage/proc.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +493,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..c399e6c267 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,68 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver ", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +144,48 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +210,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +248,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +270,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +656,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..34c8551288 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3141,24 +3142,6 @@ reaper(SIGNAL_ARGS) continue; } - /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and - * PostmasterStateMachine() will advance to the next shutdown step. - */ - if (pid == PgArchPID) - { - PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); - if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); - continue; - } - /* * Was it the statistics collector? If so, just try to start a new * one; no need to force reset of the rest of the system. (If fail, @@ -3175,6 +3158,26 @@ reaper(SIGNAL_ARGS) continue; } + /* + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and + * PostmasterStateMachine() will advance to the next shutdown step. + */ + if (pid == PgArchPID) + { + PgArchPID = 0; + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); + if (PgArchStartupAllowed()) + PgArchPID = StartArchiver(); + continue; + } + /* Was it the system logger? If so, try to start a new one */ if (pid == SysLoggerPID) { @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 1bdc97e308..adb9f819bb 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -419,6 +419,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -431,6 +432,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; -- 2.27.0 ----Next_Part(Wed_Mar_10_17_51_37_2021_192)-- Content-Type: Text/X-Patch; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline; filename="v52-0004-Shared-memory-based-stats-collector.patch" ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v53 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 227 ++++++++--------------- src/backend/postmaster/postmaster.c | 74 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 13 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 137 insertions(+), 218 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..818c9a1f3f 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,70 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + Latch *latch; /* latch to wake the archiver up */ + slock_t mutex; /* locks this struct */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) -{ - time_t curtime; - pid_t pgArchPid; + return size; +} - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif + if (!found) { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + SpinLockInit(&PgArch->mutex); + PgArch->latch = NULL; } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); } -#endif /* EXEC_BACKEND */ - -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +146,49 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = MyLatch; + SpinLockRelease(&PgArch->mutex); pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + Latch *latch; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); + /* fetching the latch pointer might not be atomic, so use spinlock */ + SpinLockAcquire(&PgArch->mutex); + latch = PgArch->latch; + SpinLockRelease(&PgArch->mutex); - errno = save_errno; + if (latch) + SetLatch(latch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +213,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +251,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +273,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +659,17 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + SpinLockAcquire(&PgArch->mutex); + PgArch->latch = NULL; + SpinLockRelease(&PgArch->mutex); +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..460e629145 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -548,6 +548,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1793,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3008,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3143,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3406,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3612,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3804,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3911,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5037,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5134,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5188,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5429,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..d053bf1564 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,9 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_09_23_13_2021_947)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v54 3/7] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 241 +++++++++-------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 160 insertions(+), 217 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..8e572afea3 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,6 +71,12 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data @@ -80,146 +84,79 @@ */ static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) +{ + Size size = 0; -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ + size = add_size(size, sizeof(PgArchData)); + + return size; +} + +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; + + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); + + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; + } +} /* - * pgarch_start + * PgArchIsSuppressed * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. + * Return true if archiver relaunch is suppressed. * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -int -pgarch_start(void) +bool +PgArchIsSuppressed(void) { - time_t curtime; - pid_t pgArchPid; + time_t curtime = time(NULL); - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; + if ((curtime - last_pgarch_start_time) < PGARCH_RESTART_INTERVAL) + return true; - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; last_pgarch_start_time = curtime; - -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); - - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); - - PgArchiverMain(0, NULL); - break; -#endif - - default: - return (int) pgArchPid; - } - - /* shouldn't get here */ - return 0; -} - -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - -/* - * pgarch_forkexec() - - * - * Format up the arglist for, then fork and exec, archive process - */ -static pid_t -pgarch_forkexec(void) -{ - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; - - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ - - av[ac] = NULL; - Assert(ac < lengthof(av)); - - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +168,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our latch that backends can use to wake us up while we're + * sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will be + * relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +237,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +275,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +297,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +683,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..4f8e364284 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + !PgArchIsSuppressed()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for it + * to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and syslogger + * processes are disregarded since they are not connected to shared + * memory; we also disregard dead_end children here. Walsenders and + * archiver are also disregarded, they will be terminated later after + * writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..640692d33e 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchIsSuppressed(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_13_49_06_2021_682)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* [PATCH v55 3/3] Make archiver process an auxiliary process @ 2021-03-11 09:01 Kyotaro Horiguchi <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: Kyotaro Horiguchi @ 2021-03-11 09:01 UTC (permalink / raw) This makes it possible to archiver process uses ipc functions and let us monitor archiver process status. --- doc/src/sgml/monitoring.sgml | 1 + src/backend/access/transam/xlogarchive.c | 4 +- src/backend/bootstrap/bootstrap.c | 22 ++- src/backend/postmaster/pgarch.c | 240 ++++++++++------------- src/backend/postmaster/postmaster.c | 81 ++++---- src/backend/storage/ipc/ipci.c | 2 + src/include/miscadmin.h | 2 + src/include/postmaster/pgarch.h | 14 +- src/include/storage/pmsignal.h | 1 - src/include/storage/proc.h | 8 +- src/tools/pgindent/typedefs.list | 1 + 11 files changed, 164 insertions(+), 212 deletions(-) diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 1ba813bbb9..a96455bdb2 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -935,6 +935,7 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser <literal>logical replication worker</literal>, <literal>parallel worker</literal>, <literal>background writer</literal>, <literal>client backend</literal>, <literal>checkpointer</literal>, + <literal>archiver</literal>, <literal>startup</literal>, <literal>walreceiver</literal>, <literal>walsender</literal> and <literal>walwriter</literal>. In addition, background workers registered by extensions may have diff --git a/src/backend/access/transam/xlogarchive.c b/src/backend/access/transam/xlogarchive.c index 1c5a4f8b5a..26b023e754 100644 --- a/src/backend/access/transam/xlogarchive.c +++ b/src/backend/access/transam/xlogarchive.c @@ -25,11 +25,11 @@ #include "common/archive.h" #include "miscadmin.h" #include "postmaster/startup.h" +#include "postmaster/pgarch.h" #include "replication/walsender.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/lwlock.h" -#include "storage/pmsignal.h" /* * Attempt to retrieve the specified file from off-line archival storage. @@ -491,7 +491,7 @@ XLogArchiveNotify(const char *xlog) /* Notify archiver that it's got something to do */ if (IsUnderPostmaster) - SendPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER); + PgArchWakeup(); } /* diff --git a/src/backend/bootstrap/bootstrap.c b/src/backend/bootstrap/bootstrap.c index 6f615e6622..41da0c5059 100644 --- a/src/backend/bootstrap/bootstrap.c +++ b/src/backend/bootstrap/bootstrap.c @@ -317,6 +317,9 @@ AuxiliaryProcessMain(int argc, char *argv[]) case StartupProcess: MyBackendType = B_STARTUP; break; + case ArchiverProcess: + MyBackendType = B_ARCHIVER; + break; case BgWriterProcess: MyBackendType = B_BG_WRITER; break; @@ -437,30 +440,29 @@ AuxiliaryProcessMain(int argc, char *argv[]) proc_exit(1); /* should never return */ case StartupProcess: - /* don't set signals, startup process has its own agenda */ StartupProcessMain(); - proc_exit(1); /* should never return */ + proc_exit(1); + + case ArchiverProcess: + PgArchiverMain(); + proc_exit(1); case BgWriterProcess: - /* don't set signals, bgwriter has its own agenda */ BackgroundWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case CheckpointerProcess: - /* don't set signals, checkpointer has its own agenda */ CheckpointerMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalWriterProcess: - /* don't set signals, walwriter has its own agenda */ InitXLOGAccess(); WalWriterMain(); - proc_exit(1); /* should never return */ + proc_exit(1); case WalReceiverProcess: - /* don't set signals, walreceiver has its own agenda */ WalReceiverMain(); - proc_exit(1); /* should never return */ + proc_exit(1); default: elog(PANIC, "unrecognized process type: %d", (int) MyAuxProcType); diff --git a/src/backend/postmaster/pgarch.c b/src/backend/postmaster/pgarch.c index edec311f12..07ba989a37 100644 --- a/src/backend/postmaster/pgarch.c +++ b/src/backend/postmaster/pgarch.c @@ -38,16 +38,14 @@ #include "libpq/pqsignal.h" #include "miscadmin.h" #include "pgstat.h" -#include "postmaster/fork_process.h" #include "postmaster/interrupt.h" #include "postmaster/pgarch.h" -#include "postmaster/postmaster.h" -#include "storage/dsm.h" #include "storage/fd.h" #include "storage/ipc.h" #include "storage/latch.h" -#include "storage/pg_shmem.h" #include "storage/pmsignal.h" +#include "storage/procsignal.h" +#include "storage/spin.h" #include "utils/guc.h" #include "utils/ps_status.h" @@ -73,153 +71,101 @@ */ #define NUM_ORPHAN_CLEANUP_RETRIES 3 +/* Shared memory area for archiver process */ +typedef struct PgArchData +{ + int pgprocno; /* pgprocno of archiver process */ +} PgArchData; + /* ---------- * Local data * ---------- */ -static time_t last_pgarch_start_time; static time_t last_sigterm_time = 0; +static PgArchData *PgArch = NULL; /* * Flags set by interrupt handlers for later service in the main loop. */ -static volatile sig_atomic_t wakened = false; static volatile sig_atomic_t ready_to_stop = false; /* ---------- * Local function forward declarations * ---------- */ -#ifdef EXEC_BACKEND -static pid_t pgarch_forkexec(void); -#endif - -NON_EXEC_STATIC void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -static void pgarch_waken(SIGNAL_ARGS); static void pgarch_waken_stop(SIGNAL_ARGS); static void pgarch_MainLoop(void); static void pgarch_ArchiverCopyLoop(void); static bool pgarch_archiveXlog(char *xlog); static bool pgarch_readyXlog(char *xlog); static void pgarch_archiveDone(char *xlog); +static void pgarch_die(int code, Datum arg); - -/* ------------------------------------------------------------ - * Public functions called from postmaster follow - * ------------------------------------------------------------ - */ - -/* - * pgarch_start - * - * Called from postmaster at startup or after an existing archiver - * died. Attempt to fire up a fresh archiver process. - * - * Returns PID of child process, or 0 if fail. - * - * Note: if fail, we will be called again from the postmaster main loop. - */ -int -pgarch_start(void) +/* Report shared memory space needed by PgArchShmemInit */ +Size +PgArchShmemSize(void) { - time_t curtime; - pid_t pgArchPid; + Size size = 0; - /* - * Do nothing if no archiver needed - */ - if (!XLogArchivingActive()) - return 0; - - /* - * Do nothing if too soon since last archiver start. This is a safety - * valve to protect against continuous respawn attempts if the archiver is - * dying immediately at launch. Note that since we will be re-called from - * the postmaster main loop, we will get another chance later. - */ - curtime = time(NULL); - if ((unsigned int) (curtime - last_pgarch_start_time) < - (unsigned int) PGARCH_RESTART_INTERVAL) - return 0; - last_pgarch_start_time = curtime; + size = add_size(size, sizeof(PgArchData)); -#ifdef EXEC_BACKEND - switch ((pgArchPid = pgarch_forkexec())) -#else - switch ((pgArchPid = fork_process())) -#endif - { - case -1: - ereport(LOG, - (errmsg("could not fork archiver: %m"))); - return 0; - -#ifndef EXEC_BACKEND - case 0: - /* in postmaster child ... */ - InitPostmasterChild(); - - /* Close the postmaster's sockets */ - ClosePostmasterPorts(false); + return size; +} - /* Drop our connection to postmaster's shared memory, as well */ - dsm_detach_all(); - PGSharedMemoryDetach(); +/* Allocate and initialize archiver-related shared memory */ +void +PgArchShmemInit(void) +{ + bool found; - PgArchiverMain(0, NULL); - break; -#endif + PgArch = (PgArchData *) + ShmemInitStruct("Archiver Data", PgArchShmemSize(), &found); - default: - return (int) pgArchPid; + if (!found) + { + /* First time through, so initialize */ + MemSet(PgArch, 0, PgArchShmemSize()); + PgArch->pgprocno = INVALID_PGPROCNO; } - - /* shouldn't get here */ - return 0; } -/* ------------------------------------------------------------ - * Local functions called by archiver follow - * ------------------------------------------------------------ - */ - - -#ifdef EXEC_BACKEND - /* - * pgarch_forkexec() - + * PgArchIsSuppressed * - * Format up the arglist for, then fork and exec, archive process + * Return true if archiver relaunch is suppressed. + * + * This is a safety valve to protect against continuous respawn attempts if the + * archiver is dying immediately at launch. Note that since we will retry to + * launch the archiver from the postmaster main loop, we will get another + * chance later. */ -static pid_t -pgarch_forkexec(void) +bool +PgArchCanRestart(void) { - char *av[10]; - int ac = 0; - - av[ac++] = "postgres"; + static time_t last_pgarch_start_time = 0; + time_t curtime = time(NULL); - av[ac++] = "--forkarch"; - - av[ac++] = NULL; /* filled in by postmaster_forkexec */ + /* + * allow to launch also for the first time and when something's wrong in + * the time values. + */ + if (last_pgarch_start_time == 0 || + curtime < last_pgarch_start_time || + (curtime - last_pgarch_start_time) >= PGARCH_RESTART_INTERVAL) + { + last_pgarch_start_time = curtime; - av[ac] = NULL; - Assert(ac < lengthof(av)); + return true; + } - return postmaster_forkexec(ac, av); + return false; } -#endif /* EXEC_BACKEND */ -/* - * PgArchiverMain - * - * The argc/argv parameters are valid only in EXEC_BACKEND case. However, - * since we don't use 'em, it hardly matters... - */ -NON_EXEC_STATIC void -PgArchiverMain(int argc, char *argv[]) +/* Main entry point for archiver process */ +void +PgArchiverMain(void) { /* * Ignore all signals usually bound to some action in the postmaster, @@ -231,33 +177,51 @@ PgArchiverMain(int argc, char *argv[]) /* SIGQUIT handler was already set up by InitPostmasterChild */ pqsignal(SIGALRM, SIG_IGN); pqsignal(SIGPIPE, SIG_IGN); - pqsignal(SIGUSR1, pgarch_waken); + pqsignal(SIGUSR1, procsignal_sigusr1_handler); pqsignal(SIGUSR2, pgarch_waken_stop); + /* Reset some signals that are accepted by postmaster but not here */ pqsignal(SIGCHLD, SIG_DFL); + + /* Unblock signals (they were blocked when the postmaster forked us) */ PG_SETMASK(&UnBlockSig); - MyBackendType = B_ARCHIVER; - init_ps_display(NULL); + /* We shouldn't be launched unnecessarily. */ + Assert(XLogArchivingActive()); + + /* Arrange to clean up at archiver exit */ + on_shmem_exit(pgarch_die, 0); + + /* + * Advertise our pgproc index so that backends can use our latch to wake + * us up while we're sleeping. + */ + PgArch->pgprocno = MyProc->pgprocno; pgarch_MainLoop(); - exit(0); + proc_exit(0); } -/* SIGUSR1 signal handler for archiver process */ -static void -pgarch_waken(SIGNAL_ARGS) +/* + * Wake up the archiver + */ +void +PgArchWakeup(void) { - int save_errno = errno; + int arch_pgprocno = PgArch->pgprocno; - /* set flag that there is work to be done */ - wakened = true; - SetLatch(MyLatch); - - errno = save_errno; + /* + * We don't acquire ProcArrayLock here. It's actually fine because + * procLatch isn't ever freed, so we just can potentially set the wrong + * process' (or no process') latch. Even in that case the archiver will + * be relaunched shortly and will start archiving. + */ + if (arch_pgprocno != INVALID_PGPROCNO) + SetLatch(&ProcGlobal->allProcs[arch_pgprocno].procLatch); } + /* SIGUSR2 signal handler for archiver process */ static void pgarch_waken_stop(SIGNAL_ARGS) @@ -282,14 +246,6 @@ pgarch_MainLoop(void) pg_time_t last_copy_time = 0; bool time_to_stop; - /* - * We run the copy loop immediately upon entry, in case there are - * unarchived files left over from a previous database run (or maybe the - * archiver died unexpectedly). After that we wait for a signal or - * timeout before doing more. - */ - wakened = true; - /* * There shouldn't be anything for the archiver to do except to wait for a * signal ... however, the archiver exists to protect our data, so she @@ -328,12 +284,8 @@ pgarch_MainLoop(void) } /* Do what we're here for */ - if (wakened || time_to_stop) - { - wakened = false; - pgarch_ArchiverCopyLoop(); - last_copy_time = time(NULL); - } + pgarch_ArchiverCopyLoop(); + last_copy_time = time(NULL); /* * Sleep until a signal is received, or until a poll is forced by @@ -354,13 +306,9 @@ pgarch_MainLoop(void) WL_LATCH_SET | WL_TIMEOUT | WL_POSTMASTER_DEATH, timeout * 1000L, WAIT_EVENT_ARCHIVER_MAIN); - if (rc & WL_TIMEOUT) - wakened = true; if (rc & WL_POSTMASTER_DEATH) time_to_stop = true; } - else - wakened = true; } /* @@ -744,3 +692,15 @@ pgarch_archiveDone(char *xlog) StatusFilePath(rlogdone, xlog, ".done"); (void) durable_rename(rlogready, rlogdone, WARNING); } + + +/* + * pgarch_die + * + * Exit-time cleanup handler + */ +static void +pgarch_die(int code, Datum arg) +{ + PgArch->pgprocno = INVALID_PGPROCNO; +} diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index edab95a19e..e8af05c04e 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -443,9 +443,10 @@ static void InitPostmasterDeathWatchHandle(void); * even during recovery. */ #define PgArchStartupAllowed() \ - ((XLogArchivingActive() && pmState == PM_RUN) || \ - (XLogArchivingAlways() && \ - (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) + (((XLogArchivingActive() && pmState == PM_RUN) || \ + (XLogArchivingAlways() && \ + (pmState == PM_RECOVERY || pmState == PM_HOT_STANDBY))) && \ + PgArchCanRestart()) #ifdef EXEC_BACKEND @@ -548,6 +549,7 @@ static void ShmemBackendArrayRemove(Backend *bn); #endif /* EXEC_BACKEND */ #define StartupDataBase() StartChildProcess(StartupProcess) +#define StartArchiver() StartChildProcess(ArchiverProcess) #define StartBackgroundWriter() StartChildProcess(BgWriterProcess) #define StartCheckpointer() StartChildProcess(CheckpointerProcess) #define StartWalWriter() StartChildProcess(WalWriterProcess) @@ -1792,7 +1794,7 @@ ServerLoop(void) /* If we have lost the archiver, try to start a new one. */ if (PgArchPID == 0 && PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* If we need to signal the autovacuum launcher, do so now */ if (avlauncher_needs_signal) @@ -3007,7 +3009,7 @@ reaper(SIGNAL_ARGS) if (!IsBinaryUpgrade && AutoVacuumingActive() && AutoVacPID == 0) AutoVacPID = StartAutoVacLauncher(); if (PgArchStartupAllowed() && PgArchPID == 0) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); if (PgStatPID == 0) PgStatPID = pgstat_start(); @@ -3142,20 +3144,22 @@ reaper(SIGNAL_ARGS) } /* - * Was it the archiver? If so, just try to start a new one; no need - * to force reset of the rest of the system. (If fail, we'll try - * again in future cycles of the main loop.). Unless we were waiting - * for it to shut down; don't restart it in that case, and + * Was it the archiver? If exit status is zero (normal) or one (FATAL + * exit), we assume everything is all right just like normal backends + * and just try to restart a new one so that we immediately retry + * archiving remaining files. (If fail, we'll try again in future + * cycles of the postmaster's main loop.) Unless we were waiting for + * it to shut down; don't restart it in that case, and * PostmasterStateMachine() will advance to the next shutdown step. */ if (pid == PgArchPID) { PgArchPID = 0; - if (!EXIT_STATUS_0(exitstatus)) - LogChildExit(LOG, _("archiver process"), - pid, exitstatus); + if (!EXIT_STATUS_0(exitstatus) && !EXIT_STATUS_1(exitstatus)) + HandleChildCrash(pid, exitstatus, + _("archiver process")); if (PgArchStartupAllowed()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); continue; } @@ -3403,7 +3407,7 @@ CleanupBackend(int pid, /* * HandleChildCrash -- cleanup after failed backend, bgwriter, checkpointer, - * walwriter, autovacuum, or background worker. + * walwriter, autovacuum, archiver or background worker. * * The objectives here are to clean up our local state about the child * process, and to signal all other remaining children to quickdie. @@ -3609,19 +3613,16 @@ HandleChildCrash(int pid, int exitstatus, const char *procname) signal_child(AutoVacPID, (SendStop ? SIGSTOP : SIGQUIT)); } - /* - * Force a power-cycle of the pgarch process too. (This isn't absolutely - * necessary, but it seems like a good idea for robustness, and it - * simplifies the state-machine logic in the case where a shutdown request - * arrives during crash processing.) - */ - if (PgArchPID != 0 && take_action) + /* Take care of the archiver too */ + if (pid == PgArchPID) + PgArchPID = 0; + else if (PgArchPID != 0 && take_action) { ereport(DEBUG2, (errmsg_internal("sending %s to process %d", - "SIGQUIT", + (SendStop ? "SIGSTOP" : "SIGQUIT"), (int) PgArchPID))); - signal_child(PgArchPID, SIGQUIT); + signal_child(PgArchPID, (SendStop ? SIGSTOP : SIGQUIT)); } /* @@ -3804,12 +3805,11 @@ PostmasterStateMachine(void) * (including autovac workers), no bgworkers (including unconnected * ones), and no walwriter, autovac launcher or bgwriter. If we are * doing crash recovery or an immediate shutdown then we expect the - * checkpointer to exit as well, otherwise not. The archiver, stats, - * and syslogger processes are disregarded since they are not - * connected to shared memory; we also disregard dead_end children - * here. Walsenders are also disregarded, they will be terminated - * later after writing the checkpoint record, like the archiver - * process. + * checkpointer to exit as well, otherwise not. The stats and + * syslogger processes are disregarded since they are not connected to + * shared memory; we also disregard dead_end children here. Walsenders + * and archiver are also disregarded, they will be terminated later + * after writing the checkpoint record. */ if (CountChildren(BACKEND_TYPE_ALL - BACKEND_TYPE_WALSND) == 0 && StartupPID == 0 && @@ -3912,6 +3912,7 @@ PostmasterStateMachine(void) Assert(CheckpointerPID == 0); Assert(WalWriterPID == 0); Assert(AutoVacPID == 0); + Assert(PgArchPID == 0); /* syslogger is not considered here */ pmState = PM_NO_CHILDREN; } @@ -5037,12 +5038,6 @@ SubPostmasterMain(int argc, char *argv[]) StartBackgroundWorker(); } - if (strcmp(argv[1], "--forkarch") == 0) - { - /* Do not want to attach to shared memory */ - - PgArchiverMain(argc, argv); /* does not return */ - } if (strcmp(argv[1], "--forkcol") == 0) { /* Do not want to attach to shared memory */ @@ -5140,7 +5135,7 @@ sigusr1_handler(SIGNAL_ARGS) */ Assert(PgArchPID == 0); if (XLogArchivingAlways()) - PgArchPID = pgarch_start(); + PgArchPID = StartArchiver(); /* * If we aren't planning to enter hot standby mode later, treat @@ -5194,16 +5189,6 @@ sigusr1_handler(SIGNAL_ARGS) if (StartWorkerNeeded || HaveCrashedWorker) maybe_start_bgworkers(); - if (CheckPostmasterSignal(PMSIGNAL_WAKEN_ARCHIVER) && - PgArchPID != 0) - { - /* - * Send SIGUSR1 to archiver process, to wake it up and begin archiving - * next WAL file. - */ - signal_child(PgArchPID, SIGUSR1); - } - /* Tell syslogger to rotate logfile if requested */ if (SysLoggerPID != 0) { @@ -5445,6 +5430,10 @@ StartChildProcess(AuxProcType type) ereport(LOG, (errmsg("could not fork startup process: %m"))); break; + case ArchiverProcess: + ereport(LOG, + (errmsg("could not fork archiver process: %m"))); + break; case BgWriterProcess: ereport(LOG, (errmsg("could not fork background writer process: %m"))); diff --git a/src/backend/storage/ipc/ipci.c b/src/backend/storage/ipc/ipci.c index f9bbe97b50..3e4ec53a97 100644 --- a/src/backend/storage/ipc/ipci.c +++ b/src/backend/storage/ipc/ipci.c @@ -144,6 +144,7 @@ CreateSharedMemoryAndSemaphores(void) size = add_size(size, ReplicationOriginShmemSize()); size = add_size(size, WalSndShmemSize()); size = add_size(size, WalRcvShmemSize()); + size = add_size(size, PgArchShmemSize()); size = add_size(size, ApplyLauncherShmemSize()); size = add_size(size, SnapMgrShmemSize()); size = add_size(size, BTreeShmemSize()); @@ -258,6 +259,7 @@ CreateSharedMemoryAndSemaphores(void) ReplicationOriginShmemInit(); WalSndShmemInit(); WalRcvShmemInit(); + PgArchShmemInit(); ApplyLauncherShmemInit(); /* diff --git a/src/include/miscadmin.h b/src/include/miscadmin.h index 54693e047a..013850ac28 100644 --- a/src/include/miscadmin.h +++ b/src/include/miscadmin.h @@ -417,6 +417,7 @@ typedef enum BootstrapProcess, StartupProcess, BgWriterProcess, + ArchiverProcess, CheckpointerProcess, WalWriterProcess, WalReceiverProcess, @@ -429,6 +430,7 @@ extern AuxProcType MyAuxProcType; #define AmBootstrapProcess() (MyAuxProcType == BootstrapProcess) #define AmStartupProcess() (MyAuxProcType == StartupProcess) #define AmBackgroundWriterProcess() (MyAuxProcType == BgWriterProcess) +#define AmArchiverProcess() (MyAuxProcType == ArchiverProcess) #define AmCheckpointerProcess() (MyAuxProcType == CheckpointerProcess) #define AmWalWriterProcess() (MyAuxProcType == WalWriterProcess) #define AmWalReceiverProcess() (MyAuxProcType == WalReceiverProcess) diff --git a/src/include/postmaster/pgarch.h b/src/include/postmaster/pgarch.h index d102a21ab7..1e47a143e1 100644 --- a/src/include/postmaster/pgarch.h +++ b/src/include/postmaster/pgarch.h @@ -26,14 +26,10 @@ #define MAX_XFN_CHARS 40 #define VALID_XFN_CHARS "0123456789ABCDEF.history.backup.partial" -/* ---------- - * Functions called from postmaster - * ---------- - */ -extern int pgarch_start(void); - -#ifdef EXEC_BACKEND -extern void PgArchiverMain(int argc, char *argv[]) pg_attribute_noreturn(); -#endif +extern Size PgArchShmemSize(void); +extern void PgArchShmemInit(void); +extern bool PgArchCanRestart(void); +extern void PgArchiverMain(void) pg_attribute_noreturn(); +extern void PgArchWakeup(void); #endif /* _PGARCH_H */ diff --git a/src/include/storage/pmsignal.h b/src/include/storage/pmsignal.h index dbbed18f61..8ed4d87ae6 100644 --- a/src/include/storage/pmsignal.h +++ b/src/include/storage/pmsignal.h @@ -34,7 +34,6 @@ typedef enum { PMSIGNAL_RECOVERY_STARTED, /* recovery has started */ PMSIGNAL_BEGIN_HOT_STANDBY, /* begin Hot Standby */ - PMSIGNAL_WAKEN_ARCHIVER, /* send a NOTIFY signal to xlog archiver */ PMSIGNAL_ROTATE_LOGFILE, /* send SIGUSR1 to syslogger to rotate logfile */ PMSIGNAL_START_AUTOVAC_LAUNCHER, /* start an autovacuum launcher */ PMSIGNAL_START_AUTOVAC_WORKER, /* start an autovacuum worker */ diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index a777cb64a1..2fd1ff09a7 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -370,11 +370,11 @@ extern PGPROC *PreparedXactProcs; * We set aside some extra PGPROC structures for auxiliary processes, * ie things that aren't full-fledged backends but need shmem access. * - * Background writer, checkpointer and WAL writer run during normal operation. - * Startup process and WAL receiver also consume 2 slots, but WAL writer is - * launched only after startup has exited, so we only need 4 slots. + * Background writer, checkpointer, WAL writer and archiver run during normal + * operation. Startup process and WAL receiver also consume 2 slots, but WAL + * writer is launched only after startup has exited, so we only need 5 slots. */ -#define NUM_AUXILIARY_PROCS 4 +#define NUM_AUXILIARY_PROCS 5 /* configurable options */ extern PGDLLIMPORT int DeadlockTimeout; diff --git a/src/tools/pgindent/typedefs.list b/src/tools/pgindent/typedefs.list index 99a34be465..78e852569e 100644 --- a/src/tools/pgindent/typedefs.list +++ b/src/tools/pgindent/typedefs.list @@ -1572,6 +1572,7 @@ PGresAttValue PGresParamDesc PGresult PGresult_data +PgArchData PHANDLE PLAINTREE PLUID_AND_ATTRIBUTES -- 2.27.0 ----Next_Part(Fri_Mar_12_17_24_43_2021_866)---- ^ permalink raw reply [nested|flat] 279+ messages in thread
* Re: Add tracking of backend memory allocated to pg_stat_activity @ 2023-01-05 18:58 Reid Thompson <[email protected]> 0 siblings, 1 reply; 279+ messages in thread From: Reid Thompson @ 2023-01-05 18:58 UTC (permalink / raw) To: vignesh C <[email protected]>; +Cc: [email protected]; Justin Pryzby <[email protected]>; Andres Freund <[email protected]>; [email protected]; Drouvot, Bertrand <[email protected]>; Melanie Plageman <[email protected]> On Tue, 2023-01-03 at 16:25 +0530, vignesh C wrote: > ... > The patch does not apply on top of HEAD as in [1], please post a > rebased patch: >... > Regards, > Vignesh Per conversation in thread listed below, patches have been submitted to the "Add the ability to limit the amount of memory that can be allocated to backends" thread https://www.postgresql.org/message-id/[email protected] 0001-Add-tracking-of-backend-memory-allocated-to-pg_stat_.patch 0002-Add-the-ability-to-limit-the-amount-of-memory-that-c.patch On Thu, 8 Dec 2022 at 19:44, Reid Thompson <reid(dot)thompson(at)crunchydata(dot)com> wrote: > > On Sun, 2022-11-27 at 09:40 -0600, Justin Pryzby wrote: > > > ... > > > I still wonder whether there needs to be a separate CF entry for > > > the 0001 patch. One issue is that there's two different lists of > > > people involved in the threads. > > > > > I'm OK with containing the conversation to one thread if everyone else > is. If there's no argument against, then patches after today will go > to the "Add the ability to limit the amount of memory that can be > allocated to backends" thread > https://www.postgresql.org/message-id/[email protected] -- Reid Thompson Senior Software Engineer Crunchy Data, Inc. [email protected] www.crunchydata.com ^ permalink raw reply [nested|flat] 279+ messages in thread
* Re: Add tracking of backend memory allocated to pg_stat_activity @ 2023-01-05 20:13 Justin Pryzby <[email protected]> parent: Reid Thompson <[email protected]> 0 siblings, 1 reply; 279+ messages in thread From: Justin Pryzby @ 2023-01-05 20:13 UTC (permalink / raw) To: Reid Thompson <[email protected]>; +Cc: vignesh C <[email protected]>; Andres Freund <[email protected]>; [email protected]; Drouvot, Bertrand <[email protected]>; Melanie Plageman <[email protected]> On Thu, Jan 05, 2023 at 01:58:33PM -0500, Reid Thompson wrote: > On Tue, 2023-01-03 at 16:25 +0530, vignesh C wrote: > > ... > > The patch does not apply on top of HEAD as in [1], please post a > > rebased patch: > >... > > Regards, > > Vignesh > > Per conversation in thread listed below, patches have been submitted to the "Add the ability to limit the amount of memory that can be allocated to backends" thread > https://www.postgresql.org/message-id/[email protected] I suggest to close the associated CF entry. (Also, the people who participated in this thread may want to be included in the other thread going forward.) > 0001-Add-tracking-of-backend-memory-allocated-to-pg_stat_.patch > 0002-Add-the-ability-to-limit-the-amount-of-memory-that-c.patch > > On Thu, 8 Dec 2022 at 19:44, Reid Thompson > <reid(dot)thompson(at)crunchydata(dot)com> wrote: > > > > On Sun, 2022-11-27 at 09:40 -0600, Justin Pryzby wrote: > > > > ... > > > > I still wonder whether there needs to be a separate CF entry for > > > > the 0001 patch. One issue is that there's two different lists of > > > > people involved in the threads. > > > > > > > > I'm OK with containing the conversation to one thread if everyone else > > is. If there's no argument against, then patches after today will go > > to the "Add the ability to limit the amount of memory that can be > > allocated to backends" thread > > https://www.postgresql.org/message-id/[email protected] ^ permalink raw reply [nested|flat] 279+ messages in thread
* Re: Add tracking of backend memory allocated to pg_stat_activity @ 2023-01-05 21:33 Reid Thompson <[email protected]> parent: Justin Pryzby <[email protected]> 0 siblings, 1 reply; 279+ messages in thread From: Reid Thompson @ 2023-01-05 21:33 UTC (permalink / raw) To: Justin Pryzby <[email protected]>; +Cc: [email protected]; vignesh C <[email protected]>; Andres Freund <[email protected]>; [email protected]; Drouvot, Bertrand <[email protected]>; Melanie Plageman <[email protected]>; Kyotaro Horiguchi <[email protected]>; StephenFrost <[email protected]>; Ted Yu <[email protected]> On Thu, 2023-01-05 at 14:13 -0600, Justin Pryzby wrote: > > I suggest to close the associated CF entry. Closed with status of Withdrawn. If that is invalid, please advise. > (Also, the people who participated in this thread may want to be > included in the other thread going forward.) Explicitly adding previously missed participants to Cc: - that conversation/patches are being consolidated to the thread "Add the ability to limit the amount of memory that can be allocated to backends" https://www.postgresql.org/message-id/[email protected] Thanks, Reid -- Reid Thompson Senior Software Engineer Crunchy Data, Inc. [email protected] www.crunchydata.com ^ permalink raw reply [nested|flat] 279+ messages in thread
* Re: Add tracking of backend memory allocated to pg_stat_activity @ 2023-08-31 16:18 John Morris <[email protected]> parent: Reid Thompson <[email protected]> 0 siblings, 0 replies; 279+ messages in thread From: John Morris @ 2023-08-31 16:18 UTC (permalink / raw) To: [email protected] <[email protected]>; Justin Pryzby <[email protected]>; +Cc: vignesh C <[email protected]>; Andres Freund <[email protected]>; [email protected] <[email protected]>; Drouvot, Bertrand <[email protected]>; Melanie Plageman <[email protected]>; Kyotaro Horiguchi <[email protected]>; StephenFrost <[email protected]>; Ted Yu <[email protected]> Here is an updated version of the earlier work. This version: 1) Tracks memory as requested by the backend. 2) Includes allocations made during program startup. 3) Optimizes the "fast path" to only update two local variables. 4) Places a cluster wide limit on total memory allocated. The cluster wide limit is useful for multi-hosting. One greedy cluster doesn't starve the other clusters of memory. Note there isn't a good way to track actual memory used by a cluster. Ideally, we like to get the working set size of each memory segment along with the size of the associated kernel data structures. Gathering that info in a portable way is a "can of worms". Instead, we're managing memory as requested by the application. While not identical, the two approaches are strongly correlated. The memory model used is 1) Each process is assumed to use a certain amount of memory simply by existing. 2) All pg memory allocations are counted, including those before the process is fully initialized. 3) Each process maintains its own local counters. These are the "truth". 4) Periodically, - local counters are added into the global, shared memory counters. - pgstats is updated - total memory is checked. For efficiency, the global total is an approximation, not a precise number. It can be off by as much as 1 MB per process. Memory limiting doesn't need precision, just a consistent and reasonable approximation. Repeating the earlier benchmark test, there is no measurable loss of performance. Attachments: [application/octet-stream] memtrack-v4.patch (107.8K, ../../BYAPR13MB267710BCC29F13F8BE97E643A0E5A@BYAPR13MB2677.namprd13.prod.outlook.com/2-memtrack-v4.patch) download | inline diff: From 39e23b34cdfa19c0611e15bf8729086a25437714 Mon Sep 17 00:00:00 2001 From: John Morris <[email protected]> Date: Wed, 30 Aug 2023 21:21:06 -0700 Subject: [PATCH] Rebasing memtrack changes onto recent main branch. --- doc/src/sgml/config.sgml | 30 ++ doc/src/sgml/monitoring.sgml | 276 ++++++++++ src/backend/catalog/system_views.sql | 38 +- src/backend/postmaster/fork_process.c | 7 + src/backend/postmaster/postmaster.c | 6 + src/backend/storage/ipc/dsm.c | 10 +- src/backend/storage/ipc/dsm_impl.c | 46 +- src/backend/storage/lmgr/proc.c | 47 ++ src/backend/utils/activity/Makefile | 1 + src/backend/utils/activity/backend_status.c | 73 ++- src/backend/utils/activity/memtrack.c | 142 +++++ src/backend/utils/activity/meson.build | 1 + src/backend/utils/adt/pgstatfuncs.c | 89 ++++ src/backend/utils/misc/guc_tables.c | 12 + src/backend/utils/misc/postgresql.conf.sample | 3 + src/backend/utils/mmgr/aset.c | 24 +- src/backend/utils/mmgr/generation.c | 21 +- src/backend/utils/mmgr/slab.c | 23 +- src/include/catalog/pg_proc.dat | 17 + src/include/storage/proc.h | 4 + src/include/utils/backend_status.h | 8 + src/include/utils/memtrack.h | 288 ++++++++++ src/test/modules/meson.build | 1 + src/test/modules/test_memtrack/.gitignore | 4 + src/test/modules/test_memtrack/Makefile | 25 + src/test/modules/test_memtrack/README | 36 ++ .../test_memtrack/expected/test_memtrack.out | 80 +++ src/test/modules/test_memtrack/meson.build | 38 ++ .../test_memtrack/sql/test_memtrack.sql | 31 ++ .../test_memtrack/test_memtrack--1.0.sql | 20 + .../modules/test_memtrack/test_memtrack.c | 496 ++++++++++++++++++ .../modules/test_memtrack/test_memtrack.conf | 1 + .../test_memtrack/test_memtrack.control | 4 + src/test/modules/test_memtrack/worker_pool.c | 416 +++++++++++++++ src/test/modules/test_memtrack/worker_pool.h | 69 +++ src/test/regress/expected/rules.out | 29 + src/test/regress/expected/stats.out | 36 ++ src/test/regress/sql/stats.sql | 20 + 38 files changed, 2437 insertions(+), 35 deletions(-) create mode 100644 src/backend/utils/activity/memtrack.c create mode 100644 src/include/utils/memtrack.h create mode 100644 src/test/modules/test_memtrack/.gitignore create mode 100644 src/test/modules/test_memtrack/Makefile create mode 100644 src/test/modules/test_memtrack/README create mode 100644 src/test/modules/test_memtrack/expected/test_memtrack.out create mode 100644 src/test/modules/test_memtrack/meson.build create mode 100644 src/test/modules/test_memtrack/sql/test_memtrack.sql create mode 100644 src/test/modules/test_memtrack/test_memtrack--1.0.sql create mode 100644 src/test/modules/test_memtrack/test_memtrack.c create mode 100644 src/test/modules/test_memtrack/test_memtrack.conf create mode 100644 src/test/modules/test_memtrack/test_memtrack.control create mode 100644 src/test/modules/test_memtrack/worker_pool.c create mode 100644 src/test/modules/test_memtrack/worker_pool.h diff --git a/doc/src/sgml/config.sgml b/doc/src/sgml/config.sgml index 694d667bf9..0df7016c3f 100644 --- a/doc/src/sgml/config.sgml +++ b/doc/src/sgml/config.sgml @@ -2145,6 +2145,36 @@ include_dir 'conf.d' </listitem> </varlistentry> + <varlistentry id="guc-max-total-backend-memory" xreflabel="max_total_backend_memory"> + <term><varname>max_total_backend_memory</varname> (<type>integer</type>) + <indexterm> + <primary><varname>max_total_backend_memory</varname> configuration parameter</primary> + </indexterm> + </term> + <listitem> + <para> + Specifies a limit to the amount of memory (MB) that may be allocated to + backends in total (i.e. this is not a per user or per backend limit). + If unset, or set to 0 it is disabled. At databse startup + max_total_backend_memory is reduced by shared_memory_size_mb + (includes shared buffers and other memory required for initialization). + Each backend process is intialized with a 1MB local allowance which + also reduces total_bkend_mem_bytes_available. Keep this in mind when + setting this value. A backend request that would exhaust the limit will + be denied with an out of memory error causing that backend's current + query/transaction to fail. Further requests will not be allocated until + dropping below the limit. This limit does not affect auxiliary backend + processes + <xref linkend="glossary-auxiliary-proc"/> or the postmaster process. + Backend memory allocations (<varname>allocated_bytes</varname>) are + displayed in the + <link linkend="monitoring-pg-stat-memory-allocation-view"><structname>pg_stat_memory_allocation</structname></link> + view. Due to the dynamic nature of memory allocations, this limit is + not exact. + </para> + </listitem> + </varlistentry> + </variablelist> </sect2> diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml index 4ff415d6a0..98084a9055 100644 --- a/doc/src/sgml/monitoring.sgml +++ b/doc/src/sgml/monitoring.sgml @@ -4522,6 +4522,282 @@ description | Waiting for a newly initialized WAL file to reach durable storage </sect2> + <sect2 id="monitoring-pg-stat-memory-allocation-view"> + <title><structname>pg_stat_memory_allocation</structname></title> + + <indexterm> + <primary>pg_stat_memory_allocation</primary> + </indexterm> + + <para> + The <structname>pg_stat_memory_allocation</structname> view will have one + row per server process, showing information related to the current memory + allocation of that process in total and by allocator type. Due to the + dynamic nature of memory allocations the allocated bytes values may not be + exact but should be sufficient for the intended purposes. Dynamic shared + memory allocations are included only in the value displayed for the backend + that created them, they are not included in the value for backends that are + attached to them to avoid double counting. Use + <function>pg_size_pretty</function> described in + <xref linkend="functions-admin-dbsize"/> to make these values more easily + readable. + </para> + + <table id="pg-stat-memory-allocation-view" xreflabel="pg_stat_memory_allocation"> + <title><structname>pg_stat_memory_allocation</structname> View</title> + <tgroup cols="1"> + <thead> + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + Column Type + </para> + <para> + Description + </para></entry> + </row> + </thead> + + <tbody> + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>datid</structfield> <type>oid</type> + </para> + <para> + OID of the database this backend is connected to + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>pid</structfield> <type>integer</type> + </para> + <para> + Process ID of this backend + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Memory currently allocated to this backend in bytes. This is the balance + of bytes allocated and freed by this backend. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>aset_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Memory currently allocated to this backend in bytes via the allocation + set allocator. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>dsm_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Memory currently allocated to this backend in bytes via the dynamic + shared memory allocator. Upon process exit, dsm allocations that have + not been freed are considered long lived and added to + <structfield>global_dsm_allocated_bytes</structfield> found in the + <link linkend="monitoring-pg-stat-global-memory-allocation-view"> + <structname>pg_stat_global_memory_allocation</structname></link> view. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>generation_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Memory currently allocated to this backend in bytes via the generation + allocator. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>slab_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Memory currently allocated to this backend in bytes via the slab + allocator. + </para></entry> + </row> + + </tbody> + </tgroup> + </table> + + </sect2> + + <sect2 id="monitoring-pg-stat-global-memory-allocation-view"> + <title><structname>pg_stat_global_memory_allocation</structname></title> + + <indexterm> + <primary>pg_stat_global-memory_allocation</primary> + </indexterm> + + <para> + The <structname>pg_stat_global_memory_allocation</structname> view will + have one row showing information related to current shared memory + allocations. Due to the dynamic nature of memory allocations the allocated + bytes values may not be exact but should be sufficient for the intended + purposes. Use <function>pg_size_pretty</function> described in + <xref linkend="functions-admin-dbsize"/> to make the byte populated values + more easily readable. + </para> + + <table id="pg-stat-global-memory-allocation-view" xreflabel="pg_stat_global_memory_allocation"> + <title><structname>pg_stat_global_memory_allocation</structname> View</title> + <tgroup cols="1"> + <thead> + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + Column Type + </para> + <para> + Description + </para></entry> + </row> + </thead> + + <tbody> + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>datid</structfield> <type>oid</type> + </para> + <para> + OID of the database this backend is connected to + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>shared_memory_size_mb</structfield> <type>integer</type> + </para> + <para> + Reports the size of the main shared memory area, rounded up to the + nearest megabyte. See <xref linkend="guc-shared-memory-size"/>. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>shared_memory_size_in_huge_pages</structfield> <type>bigint</type> + </para> + <para> + Reports the number of huge pages that are needed for the main shared + memory area based on the specified huge_page_size. If huge pages are not + supported, this will be -1. See + <xref linkend="guc-shared-memory-size-in-huge-pages"/>. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>max_total_backend_memory_bytes</structfield> <type>bigint</type> + </para> + <para> + Reports the user defined backend maximum allowed shared memory in bytes. + 0 if disabled or not set. See + <xref linkend="guc-max-total-backend-memory"/>. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>total_bkend_mem_bytes_available</structfield> <type>bigint</type> + </para> + <para> + Tracks max_total_backend_memory (in bytes) available for allocation. At + database startup, total_bkend_mem_bytes_available is reduced by the + byte equivalent of shared_memory_size_mb. Each backend process is + intialized with a 1MB local allowance which also reduces + total_bkend_mem_bytes_available. A process's allocation requests reduce + it's local allowance. If a process's allocation request exceeds it's + remaining allowance, an attempt is made to refill the local allowance + from total_bkend_mem_bytes_available. If the refill request fails, then + the requesting process will fail with an out of memory error resulting + in the cancellation of that process's active query/transaction. The + default refill allocation quantity is 1MB. If a request is greater than + 1MB, an attempt will be made to allocate the full amount. If + max_total_backend_memory is disabled, this will be -1. + <xref linkend="guc-max-total-backend-memory"/>. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>global_dsm_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Long lived dynamically allocated memory currently allocated to the + database. Upon process exit, dsm allocations that have not been freed + are considered long lived and added to + <structfield>global_dsm_allocated_bytes</structfield>. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>total_aset_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Sum total of <structfield>aset_allocated_bytes</structfield> for all + backend processes from + <link linkend="monitoring-pg-stat-memory-allocation-view"> + <structname>pg_stat_memory_allocation</structname></link> view. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>total_dsm_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Sum total of <structfield>dsm_allocated_bytes</structfield> for all + backend processes from + <link linkend="monitoring-pg-stat-memory-allocation-view"> + <structname>pg_stat_memory_allocation</structname></link> view. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>total_generation_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Sum total of <structfield>generation_allocated_bytes</structfield> for + all backend processes from + <link linkend="monitoring-pg-stat-memory-allocation-view"> + <structname>pg_stat_memory_allocation</structname></link> view. + </para></entry> + </row> + + <row> + <entry role="catalog_table_entry"><para role="column_definition"> + <structfield>total_slab_allocated_bytes</structfield> <type>bigint</type> + </para> + <para> + Sum total of <structfield>slab_allocated_bytes</structfield> for all + backend processes from + <link linkend="monitoring-pg-stat-memory-allocation-view"> + <structname>pg_stat_memory_allocation</structname></link> view. + </para></entry> + </row> + + </tbody> + </tgroup> + </table> + + </sect2> + <sect2 id="monitoring-stats-functions"> <title>Statistics Functions</title> diff --git a/src/backend/catalog/system_views.sql b/src/backend/catalog/system_views.sql index 77b06e2a7a..364ead785e 100644 --- a/src/backend/catalog/system_views.sql +++ b/src/backend/catalog/system_views.sql @@ -1343,5 +1343,41 @@ CREATE VIEW pg_stat_subscription_stats AS FROM pg_subscription as s, pg_stat_get_subscription_stats(s.oid) as ss; +CREATE VIEW pg_stat_memory_allocation AS + SELECT + S.datid AS datid, + S.pid, + S.allocated_bytes, + S.aset_allocated_bytes, + S.dsm_allocated_bytes, + S.generation_allocated_bytes, + S.slab_allocated_bytes + FROM pg_stat_get_memory_allocation(NULL) AS S + LEFT JOIN pg_database AS D ON (S.datid = D.oid); + CREATE VIEW pg_wait_events AS - SELECT * FROM pg_get_wait_events(); + SELECT * FRom pg_get_wait_events(); + +CREATE VIEW pg_stat_global_memory_allocation AS +WITH sums AS ( + SELECT + SUM(aset_allocated_bytes) AS total_aset_allocated_bytes, + SUM(dsm_allocated_bytes) AS total_dsm_allocated_bytes, + SUM(generation_allocated_bytes) AS total_generation_allocated_bytes, + SUM(slab_allocated_bytes) AS total_slab_allocated_bytes + FROM + pg_stat_memory_allocation +) +SELECT + S.datid AS datid, + current_setting('shared_memory_size', true) as shared_memory_size, + (current_setting('shared_memory_size_in_huge_pages', true))::integer as shared_memory_size_in_huge_pages, + pg_size_bytes(current_setting('max_total_backend_memory', true)) as max_total_backend_memory_bytes, + S.total_bkend_mem_bytes_available, + S.global_dsm_allocated_bytes, + sums.total_aset_allocated_bytes, + sums.total_dsm_allocated_bytes, + sums.total_generation_allocated_bytes, + sums.total_slab_allocated_bytes + FROM sums, pg_stat_get_global_memory_allocation() AS S + LEFT JOIN pg_database AS D ON (S.datid = D.oid); diff --git a/src/backend/postmaster/fork_process.c b/src/backend/postmaster/fork_process.c index 6f9c2765d6..b7fbd31e4f 100644 --- a/src/backend/postmaster/fork_process.c +++ b/src/backend/postmaster/fork_process.c @@ -20,6 +20,7 @@ #include "libpq/pqsignal.h" #include "postmaster/fork_process.h" +#include "utils/memtrack.h" #ifndef WIN32 /* @@ -111,6 +112,12 @@ fork_process(void) } } + /* + * Reset memory tracking to the initial state. + * (They are corrrect on exec() but not after fork()) + */ + init_backend_memory(); + /* do post-fork initialization for random number generation */ pg_strong_random_init(); } diff --git a/src/backend/postmaster/postmaster.c b/src/backend/postmaster/postmaster.c index d7bfb28ff3..794bec2672 100644 --- a/src/backend/postmaster/postmaster.c +++ b/src/backend/postmaster/postmaster.c @@ -538,6 +538,7 @@ typedef struct #endif char my_exec_path[MAXPGPATH]; char pkglib_path[MAXPGPATH]; + int max_total_bkend_mem; } BackendParameters; static void read_backend_variables(char *id, Port *port); @@ -6097,6 +6098,8 @@ save_backend_variables(BackendParameters *param, Port *port, strlcpy(param->pkglib_path, pkglib_path, MAXPGPATH); + param->max_total_bkend_mem = max_total_bkend_mem; + return true; } @@ -6325,6 +6328,9 @@ restore_backend_variables(BackendParameters *param, Port *port) strlcpy(pkglib_path, param->pkglib_path, MAXPGPATH); + max_total_bkend_mem = param->max_total_bkend_mem; + max_global_bkend_bytes = (uint64)max_total_bkend_mem * 1024 * 1024; + /* * We need to restore fd.c's counts of externally-opened FDs; to avoid * confusion, be sure to do this after restoring max_safe_fds. (Note: diff --git a/src/backend/storage/ipc/dsm.c b/src/backend/storage/ipc/dsm.c index 10b029bb16..8ef1a5d0c3 100644 --- a/src/backend/storage/ipc/dsm.c +++ b/src/backend/storage/ipc/dsm.c @@ -775,6 +775,14 @@ dsm_detach_all(void) void dsm_detach(dsm_segment *seg) { + /* + * Retain mapped_size to pass into destroy call in cases where the detach + * is the last reference. mapped_size is zeroed as part of the detach + * process, but is needed later in these cases for dsm_allocated_bytes + * accounting. + */ + Size save_mapped_size = seg->mapped_size; + /* * Invoke registered callbacks. Just in case one of those callbacks * throws a further error that brings us back here, pop the callback @@ -855,7 +863,7 @@ dsm_detach(dsm_segment *seg) */ if (is_main_region_dsm_handle(seg->handle) || dsm_impl_op(DSM_OP_DESTROY, seg->handle, 0, &seg->impl_private, - &seg->mapped_address, &seg->mapped_size, WARNING)) + &seg->mapped_address, &save_mapped_size, WARNING)) { LWLockAcquire(DynamicSharedMemoryControlLock, LW_EXCLUSIVE); if (is_main_region_dsm_handle(seg->handle)) diff --git a/src/backend/storage/ipc/dsm_impl.c b/src/backend/storage/ipc/dsm_impl.c index 35fa910d6f..437a9d3d67 100644 --- a/src/backend/storage/ipc/dsm_impl.c +++ b/src/backend/storage/ipc/dsm_impl.c @@ -160,37 +160,67 @@ dsm_impl_op(dsm_op op, dsm_handle handle, Size request_size, void **impl_private, void **mapped_address, Size *mapped_size, int elevel) { + bool success; + int64 save_mapped_size = *mapped_size; + Assert(op == DSM_OP_CREATE || request_size == 0); Assert((op != DSM_OP_CREATE && op != DSM_OP_ATTACH) || (*mapped_address == NULL && *mapped_size == 0)); + /* Try to reserve memory for the backend if we're creating a new segment. */ + if (op == DSM_OP_CREATE && !reserve_backend_memory(request_size, PG_ALLOC_DSM)) + { + ereport(elevel, + (errcode(ERRCODE_OUT_OF_MEMORY), + errmsg("Unable to reserve backend memory for dynamic shared memory segment."))); + return false; + } + switch (dynamic_shared_memory_type) { #ifdef USE_DSM_POSIX case DSM_IMPL_POSIX: - return dsm_impl_posix(op, handle, request_size, impl_private, - mapped_address, mapped_size, elevel); + success = dsm_impl_posix(op, handle, request_size, impl_private, + mapped_address, mapped_size, elevel); + break; #endif #ifdef USE_DSM_SYSV case DSM_IMPL_SYSV: - return dsm_impl_sysv(op, handle, request_size, impl_private, - mapped_address, mapped_size, elevel); + success = dsm_impl_sysv(op, handle, request_size, impl_private, + mapped_address, mapped_size, elevel); + break; #endif #ifdef USE_DSM_WINDOWS case DSM_IMPL_WINDOWS: - return dsm_impl_windows(op, handle, request_size, impl_private, + success = dsm_impl_windows(op, handle, request_size, impl_private, mapped_address, mapped_size, elevel); + break #endif #ifdef USE_DSM_MMAP case DSM_IMPL_MMAP: - return dsm_impl_mmap(op, handle, request_size, impl_private, - mapped_address, mapped_size, elevel); + success = dsm_impl_mmap(op, handle, request_size, impl_private, + mapped_address, mapped_size, elevel); + break; #endif default: elog(ERROR, "unexpected dynamic shared memory type: %d", dynamic_shared_memory_type); - return false; + success = false; } + + /* CASE: we failed to create a new segment. Release the reservation */ + if ( !success && op == DSM_OP_CREATE ) + release_backend_memory(request_size, PG_ALLOC_DSM); + + /* CASE: we created a different size than expected. Record the change, but don't abort */ + else if (success && op == DSM_OP_CREATE && *mapped_size != request_size) + update_local_allocation(*mapped_size - request_size, PG_ALLOC_DSM); + + /* CASE: we destroyed a segment. Release the reservation */ + else if (success && op == DSM_OP_DESTROY) + release_backend_memory(save_mapped_size, PG_ALLOC_DSM); + + return success; } #ifdef USE_DSM_POSIX diff --git a/src/backend/storage/lmgr/proc.c b/src/backend/storage/lmgr/proc.c index 5b663a2997..d966707792 100644 --- a/src/backend/storage/lmgr/proc.c +++ b/src/backend/storage/lmgr/proc.c @@ -51,6 +51,7 @@ #include "storage/procsignal.h" #include "storage/spin.h" #include "storage/standby.h" +#include "utils/guc.h" #include "utils/timeout.h" #include "utils/timestamp.h" @@ -180,6 +181,52 @@ InitProcGlobal(void) ProcGlobal->checkpointerLatch = NULL; pg_atomic_init_u32(&ProcGlobal->procArrayGroupFirst, INVALID_PGPROCNO); pg_atomic_init_u32(&ProcGlobal->clogGroupFirst, INVALID_PGPROCNO); + pg_atomic_init_u64(&ProcGlobal->total_bkend_mem_bytes, 0); + pg_atomic_init_u64(&ProcGlobal->global_dsm_allocation, 0); + + /* Setup backend memory limiting if configured */ + if (max_total_bkend_mem > 0) + { + /* + * Convert max_total_bkend_mem to bytes, account for + * shared_memory_size, and initialize total_bkend_mem_bytes. + */ + int result = 0; + + /* Get integer value of shared_memory_size */ + if (parse_int(GetConfigOption("shared_memory_size", true, false), &result, 0, NULL)) + { + /* + * Error on startup if backend memory limit is less than shared + * memory size. Warn on startup if backend memory available is + * less than arbitrarily picked value of 100MB. + */ + + elog(WARNING, "proc init: max_total=%d result=%d\n", max_total_bkend_mem, result); + + if (max_total_bkend_mem <= result) + { + ereport(ERROR, + errmsg("configured max_total_backend_memory %dMB is <= shared_memory_size %dMB", + max_total_bkend_mem, result), + errhint("Disable or increase the configuration parameter \"max_total_backend_memory\".")); + } + else if (max_total_bkend_mem < result + 100) + { + ereport(WARNING, + errmsg("max_total_backend_memory %dMB - shared_memory_size %dMB is <= 100MB", + max_total_bkend_mem, result), + errhint("Consider increasing the configuration parameter \"max_total_backend_memory\".")); + } + + /* + * We would like to use it as bytes rather than MB. + */ + max_total_bkend_bytes = (uint64)max_total_bkend_mem * 1024 * 1024; + } + else + ereport(ERROR, errmsg("max_total_backend_memory initialization is unable to parse shared_memory_size")); + } /* * Create and initialize all the PGPROC structures we'll need. There are diff --git a/src/backend/utils/activity/Makefile b/src/backend/utils/activity/Makefile index f57cf3958c..0e0051e34a 100644 --- a/src/backend/utils/activity/Makefile +++ b/src/backend/utils/activity/Makefile @@ -18,6 +18,7 @@ override CPPFLAGS := -I. -I$(srcdir) $(CPPFLAGS) OBJS = \ backend_progress.o \ backend_status.o \ + memtrack.o \ pgstat.o \ pgstat_archiver.o \ pgstat_bgwriter.o \ diff --git a/src/backend/utils/activity/backend_status.c b/src/backend/utils/activity/backend_status.c index 722c5acf38..aa398f3ac0 100644 --- a/src/backend/utils/activity/backend_status.c +++ b/src/backend/utils/activity/backend_status.c @@ -24,6 +24,7 @@ #include "utils/backend_status.h" #include "utils/guc.h" /* for application_name */ #include "utils/memutils.h" +#include "utils/memtrack.h" /* ---------- @@ -49,7 +50,6 @@ int pgstat_track_activity_query_size = 1024; /* exposed so that backend_progress.c can access it */ PgBackendStatus *MyBEEntry = NULL; - static PgBackendStatus *BackendStatusArray = NULL; static char *BackendAppnameBuffer = NULL; static char *BackendClientHostnameBuffer = NULL; @@ -62,7 +62,6 @@ static PgBackendSSLStatus *BackendSslStatusBuffer = NULL; static PgBackendGSSStatus *BackendGssStatusBuffer = NULL; #endif - /* Status for backends including auxiliary */ static LocalPgBackendStatus *localBackendStatusTable = NULL; @@ -71,7 +70,6 @@ static int localNumBackends = 0; static MemoryContext backendStatusSnapContext; - static void pgstat_beshutdown_hook(int code, Datum arg); static void pgstat_read_current_status(void); static void pgstat_setup_backend_status_context(void); @@ -401,6 +399,9 @@ pgstat_bestart(void) lbeentry.st_progress_command_target = InvalidOid; lbeentry.st_query_id = UINT64CONST(0); + /* No allocations have been reported yet */ + lbeentry.st_memory = NO_BACKEND_MEMORY; + /* * we don't zero st_progress_param here to save cycles; nobody should * examine it until st_progress_command has been set to something other @@ -471,6 +472,9 @@ pgstat_beshutdown_hook(int code, Datum arg) PGSTAT_END_WRITE_ACTIVITY(beentry); + /* Stop reporting memory allocation changes to shared memory */ + exit_backend_memory(); + /* so that functions can check if backend_status.c is up via MyBEEntry */ MyBEEntry = NULL; } @@ -1153,6 +1157,69 @@ pgstat_get_local_beentry_by_index(int idx) return &localBackendStatusTable[idx - 1]; } +/* ---------- + * pgstat_fetch_stat_beentry() - + * + * Support function for the SQL-callable pgstat* functions. Returns + * our local copy of the current-activity entry for one backend, + * or NULL if the given beid doesn't identify any known session. + * + * The beid argument is the BackendId of the desired session + * (note that this is unlike pgstat_fetch_stat_local_beentry()). + * + * NB: caller is responsible for a check if the user is permitted to see + * this info (especially the querystring). + * ---------- + */ +PgBackendStatus * +pgstat_fetch_stat_beentry(BackendId beid) +{ + LocalPgBackendStatus key; + LocalPgBackendStatus *ret; + + pgstat_read_current_status(); + + /* + * Since the localBackendStatusTable is in order by backend_id, we can use + * bsearch() to search it efficiently. + */ + key.backend_id = beid; + ret = (LocalPgBackendStatus *) bsearch(&key, localBackendStatusTable, + localNumBackends, + sizeof(LocalPgBackendStatus), + cmp_lbestatus); + if (ret) + return &ret->backendStatus; + + return NULL; +} + + +/* ---------- + * pgstat_fetch_stat_local_beentry() - + * + * Like pgstat_fetch_stat_beentry() but with locally computed additions (like + * xid and xmin values of the backend) + * + * The beid argument is a 1-based index in the localBackendStatusTable + * (note that this is unlike pgstat_fetch_stat_beentry()). + * Returns NULL if the argument is out of range (no current caller does that). + * + * NB: caller is responsible for a check if the user is permitted to see + * this info (especially the querystring). + * ---------- + */ +LocalPgBackendStatus * +pgstat_fetch_stat_local_beentry(int beid) +{ + pgstat_read_current_status(); + + if (beid < 1 || beid > localNumBackends) + return NULL; + + return &localBackendStatusTable[beid - 1]; +} + /* ---------- * pgstat_fetch_stat_numbackends() - diff --git a/src/backend/utils/activity/memtrack.c b/src/backend/utils/activity/memtrack.c new file mode 100644 index 0000000000..73df1d916e --- /dev/null +++ b/src/backend/utils/activity/memtrack.c @@ -0,0 +1,142 @@ + +#include <unistd.h> + +#include "postgres.h" +#include "miscadmin.h" +#include "utils/backend_status.h" +#include "utils/memtrack.h" +#include "storage/proc.h" + +/* + * Max backend memory allocation allowed (MB). 0 = disabled. + * Max backend bytes is the same but in bytes. + * These default to "0", meaning don't check bounds for total memory. + */ +int max_total_bkend_mem = 0; +int64 max_total_bkend_bytes = 0; + +/* + * Local variables for tracking memory use. + * These values are preset so memory tracking is active on startup. + * After a fork(), they must be reset using 'init_backend_memory()'. + */ +PgBackendMemoryStatus my_memory = INIT_BACKEND_MEMORY; +PgBackendMemoryStatus reported_memory = NO_BACKEND_MEMORY; +int64 allocation_lower_bound = 0; +int64 allocation_upper_bound = 0; + +/* --------- + * init_backend_memory() - + * + * Called immediately after a fork(). + * Resets local memory counters to their initial startup values + */ +void +init_backend_memory(void) +{ + debug("\n"); + /* Start with nothing allocated. */ + my_memory = INIT_BACKEND_MEMORY; + reported_memory = NO_BACKEND_MEMORY; + + /* Force the next allocation to do global bounds checking. */ + allocation_lower_bound = 0; + allocation_upper_bound = 0; +} + +/* + * Clean up memory counters as backend is exiting. + * + * DSM memory is not automatically returned, so it persists in the counters. + * All other memory will disappear, so those counters are set to zero. + * + * Ideally, this function would be called last, but in practice there are some + * late memory releases that happen after it is called. + */ +void +exit_backend_memory(void) +{ + debug("\n"); + /* + * Release non-dsm memory. + * We don't release dsm shared memory since it survives process exit. + */ + for (int type = 0; type < PG_ALLOC_TYPE_MAX; type++) + if (type != PG_ALLOC_DSM) + release_backend_memory(my_memory.allocated_bytes_by_type[type], type); + + /* Force the final values to be posted to shmem */ + update_global_allocation(0, PG_ALLOC_OTHER); + + /* If we get a late request, send it to the long path. */ + allocation_lower_bound = 0; + allocation_upper_bound = 0; +} + + +/* + * Update backend memory allocation for a new request. + * + * There are two versions of this function. This one, which updates + * global values in shared memory, and an optimized update_local_allocation() + * which only updates local values. + * + * This version is the "slow path". We invoke it periodically to update + * global values and pgstat statistics. + */ +bool update_global_allocation(int64 size, pg_allocator_type type) +{ + int64 new_allocated_bytes; + int64 dsm_delta; + + /* If we are still initializing, ignore the request. It should be part of initial allocation. */ + if (ProcGlobal == NULL || MyProcPid == 0) + return update_local_allocation(size, type); + + debug("size=%ld type=%d\n", size, type); + + /* Calculate new number of bytes reflecting the reservation/release */ + new_allocated_bytes = my_memory.allocated_bytes + size; + + /* If reserving new memory and we are limited by max_total_bkend ... */ + if (size > 0 && max_total_bkend_bytes > 0 && MyAuxProcType == NotAnAuxProcess && MyProcPid != PostmasterPid) + { + /* Update the global total memory counter subject to the upper limit. */ + if (!atomic_add_within_bounds_i64(&ProcGlobal->total_bkend_mem_bytes, + new_allocated_bytes - reported_memory.allocated_bytes, + 0, max_total_bkend_bytes)) + return false; + } + + /* Otherwise, update the global counter with no limit checking */ + else + pg_atomic_add_fetch_u64(&ProcGlobal->total_bkend_mem_bytes, + new_allocated_bytes - reported_memory.allocated_bytes); + + /* Update the local memory counters. This must happen after bounds checking */ + update_local_allocation(size, type); + + /* Update the global dsm memory counter to reflect changes since our last report */ + dsm_delta = my_memory.allocated_bytes_by_type[PG_ALLOC_DSM] - reported_memory.allocated_bytes_by_type[PG_ALLOC_DSM]; + pg_atomic_add_fetch_u64(&ProcGlobal->global_dsm_allocation, dsm_delta); + + /* Update pgstat statistics if we are initialized as a backend process. */ + if (MyBEEntry != NULL) + { + PGSTAT_BEGIN_WRITE_ACTIVITY(MyBEEntry); + MyBEEntry->st_memory = my_memory; + PGSTAT_END_WRITE_ACTIVITY(MyBEEntry); + } + + /* Remember the values we just reported to pgstat */ + reported_memory = my_memory; + + /* Update bounds so they bracket our new allocation size. */ + allocation_upper_bound = my_memory.allocated_bytes + allocation_allowance_refill_qty; + allocation_lower_bound = my_memory.allocated_bytes - allocation_allowance_refill_qty; + + Assert((int64)pg_atomic_read_u64(&ProcGlobal->total_bkend_mem_bytes) >= 0); + Assert((int64)pg_atomic_read_u64(&ProcGlobal->global_dsm_allocation) >= 0); + + return true; +} diff --git a/src/backend/utils/activity/meson.build b/src/backend/utils/activity/meson.build index 46a27e7548..d3db919704 100644 --- a/src/backend/utils/activity/meson.build +++ b/src/backend/utils/activity/meson.build @@ -3,6 +3,7 @@ backend_sources += files( 'backend_progress.c', 'backend_status.c', + 'memtrack.c', 'pgstat.c', 'pgstat_archiver.c', 'pgstat_bgwriter.c', diff --git a/src/backend/utils/adt/pgstatfuncs.c b/src/backend/utils/adt/pgstatfuncs.c index dd5094a2d4..1f5aac422b 100644 --- a/src/backend/utils/adt/pgstatfuncs.c +++ b/src/backend/utils/adt/pgstatfuncs.c @@ -2067,3 +2067,92 @@ pg_stat_have_stats(PG_FUNCTION_ARGS) PG_RETURN_BOOL(pgstat_have_entry(kind, dboid, objoid)); } + +/* + * Get the memory allocation of PG backends. + */ +Datum +pg_stat_get_memory_allocation(PG_FUNCTION_ARGS) +{ +#define PG_STAT_GET_MEMORY_ALLOCATION_COLS 7 + int num_backends = pgstat_fetch_stat_numbackends(); + int curr_backend; + int pid = PG_ARGISNULL(0) ? -1 : PG_GETARG_INT32(0); + ReturnSetInfo *rsinfo = (ReturnSetInfo *) fcinfo->resultinfo; + + InitMaterializedSRF(fcinfo, 0); + + /* 1-based index */ + for (curr_backend = 1; curr_backend <= num_backends; curr_backend++) + { + /* for each row */ + Datum values[PG_STAT_GET_MEMORY_ALLOCATION_COLS] = {0}; + bool nulls[PG_STAT_GET_MEMORY_ALLOCATION_COLS] = {0}; + LocalPgBackendStatus *local_beentry; + PgBackendStatus *beentry; + + /* Get the next one in the list */ + local_beentry = pgstat_fetch_stat_local_beentry(curr_backend); + beentry = &local_beentry->backendStatus; + + /* If looking for specific PID, ignore all the others */ + if (pid != -1 && beentry->st_procpid != pid) + continue; + + /* Values available to all callers */ + if (beentry->st_databaseid != InvalidOid) + values[0] = ObjectIdGetDatum(beentry->st_databaseid); + else + nulls[0] = true; + + values[1] = Int32GetDatum(beentry->st_procpid); + values[2] = UInt64GetDatum(beentry->st_memory.allocated_bytes); + values[3] = UInt64GetDatum(beentry->st_memory.allocated_bytes_by_type[PG_ALLOC_ASET]); + values[4] = UInt64GetDatum(beentry->st_memory.allocated_bytes_by_type[PG_ALLOC_DSM]); + values[5] = UInt64GetDatum(beentry->st_memory.allocated_bytes_by_type[PG_ALLOC_GENERATION]); + values[6] = UInt64GetDatum(beentry->st_memory.allocated_bytes_by_type[PG_ALLOC_SLAB]); + + tuplestore_putvalues(rsinfo->setResult, rsinfo->setDesc, values, nulls); + + /* If only a single backend was requested, and we found it, break. */ + if (pid != -1) + break; + } + + return (Datum) 0; +} + +/* + * Get the global memory allocation statistics. + */ +Datum +pg_stat_get_global_memory_allocation(PG_FUNCTION_ARGS) +{ +#define PG_STAT_GET_GLOBAL_MEMORY_ALLOCATION_COLS 3 + TupleDesc tupdesc; + Datum values[PG_STAT_GET_GLOBAL_MEMORY_ALLOCATION_COLS] = {0}; + bool nulls[PG_STAT_GET_GLOBAL_MEMORY_ALLOCATION_COLS] = {0}; + volatile PROC_HDR *procglobal = ProcGlobal; + + /* Initialise attributes information in the tuple descriptor */ + tupdesc = CreateTemplateTupleDesc(PG_STAT_GET_GLOBAL_MEMORY_ALLOCATION_COLS); + TupleDescInitEntry(tupdesc, (AttrNumber) 1, "datid", + OIDOID, -1, 0); + TupleDescInitEntry(tupdesc, (AttrNumber) 2, "total_bkend_mem_bytes_available", + INT8OID, -1, 0); + TupleDescInitEntry(tupdesc, (AttrNumber) 3, "global_dsm_allocated_bytes", + INT8OID, -1, 0); + BlessTupleDesc(tupdesc); + + /* datid */ + values[0] = ObjectIdGetDatum(MyDatabaseId); + + /* Get total_bkend_mem_bytes */ + values[1] = Int64GetDatum(pg_atomic_read_u64(&procglobal->total_bkend_mem_bytes)); + + /* Get global_dsm_allocated_bytes */ + values[2] = Int64GetDatum(pg_atomic_read_u64(&procglobal->global_dsm_allocation)); + + /* Returns the record as Datum */ + PG_RETURN_DATUM(HeapTupleGetDatum(heap_form_tuple(tupdesc, values, nulls))); +} diff --git a/src/backend/utils/misc/guc_tables.c b/src/backend/utils/misc/guc_tables.c index e565a3092f..9d2dea3948 100644 --- a/src/backend/utils/misc/guc_tables.c +++ b/src/backend/utils/misc/guc_tables.c @@ -77,6 +77,7 @@ #include "utils/float.h" #include "utils/guc_hooks.h" #include "utils/guc_tables.h" +#include "utils/memtrack.h" #include "utils/memutils.h" #include "utils/pg_locale.h" #include "utils/portal.h" @@ -3507,6 +3508,17 @@ struct config_int ConfigureNamesInt[] = NULL, NULL, NULL }, + { + {"max_total_backend_memory", PGC_SU_BACKEND, RESOURCES_MEM, + gettext_noop("Restrict total backend memory allocations to this max."), + gettext_noop("0 turns this feature off."), + GUC_UNIT_MB + }, + &max_total_bkend_mem, + 0, 0, INT_MAX, + NULL, NULL, NULL + }, + /* End-of-list marker */ { {NULL, 0, 0, NULL, NULL}, NULL, 0, 0, 0, NULL, NULL, NULL diff --git a/src/backend/utils/misc/postgresql.conf.sample b/src/backend/utils/misc/postgresql.conf.sample index c768af9a73..bb75e2756a 100644 --- a/src/backend/utils/misc/postgresql.conf.sample +++ b/src/backend/utils/misc/postgresql.conf.sample @@ -160,6 +160,9 @@ #vacuum_buffer_usage_limit = 256kB # size of vacuum and analyze buffer access strategy ring; # 0 to disable vacuum buffer access strategy; # range 128kB to 16GB +#max_total_backend_memory = 0MB # Restrict total backend memory allocations + # to this max (in MB). 0 turns this feature + # off. # - Disk - diff --git a/src/backend/utils/mmgr/aset.c b/src/backend/utils/mmgr/aset.c index c3affaf5a8..fe8ed8a712 100644 --- a/src/backend/utils/mmgr/aset.c +++ b/src/backend/utils/mmgr/aset.c @@ -48,6 +48,7 @@ #include "port/pg_bitutils.h" #include "utils/memdebug.h" +#include "utils/memtrack.h" #include "utils/memutils.h" #include "utils/memutils_memorychunk.h" #include "utils/memutils_internal.h" @@ -441,7 +442,7 @@ AllocSetContextCreateInternal(MemoryContext parent, * Allocate the initial block. Unlike other aset.c blocks, it starts with * the context header and its block header follows that. */ - set = (AllocSet) malloc(firstBlockSize); + set = (AllocSet) malloc_backend(firstBlockSize, PG_ALLOC_ASET); if (set == NULL) { if (TopMemoryContext) @@ -539,6 +540,7 @@ AllocSetReset(MemoryContext context) AllocSet set = (AllocSet) context; AllocBlock block; Size keepersize PG_USED_FOR_ASSERTS_ONLY; + uint64 deallocation = 0; Assert(AllocSetIsValid(set)); @@ -581,6 +583,7 @@ AllocSetReset(MemoryContext context) { /* Normal case, release the block */ context->mem_allocated -= block->endptr - ((char *) block); + deallocation += block->endptr - ((char *) block); #ifdef CLOBBER_FREED_MEMORY wipe_mem(block, block->freeptr - ((char *) block)); @@ -591,6 +594,7 @@ AllocSetReset(MemoryContext context) } Assert(context->mem_allocated == keepersize); + release_backend_memory(deallocation, PG_ALLOC_ASET); /* Reset block size allocation sequence, too */ set->nextBlockSize = set->initBlockSize; @@ -609,6 +613,7 @@ AllocSetDelete(MemoryContext context) AllocSet set = (AllocSet) context; AllocBlock block = set->blocks; Size keepersize PG_USED_FOR_ASSERTS_ONLY; + uint64 deallocation = 0; Assert(AllocSetIsValid(set)); @@ -647,11 +652,13 @@ AllocSetDelete(MemoryContext context) freelist->first_free = (AllocSetContext *) oldset->header.nextchild; freelist->num_free--; + deallocation += oldset->header.mem_allocated; /* All that remains is to free the header/initial block */ free(oldset); } Assert(freelist->num_free == 0); + release_backend_memory(deallocation, PG_ALLOC_ASET); } /* Now add the just-deleted context to the freelist. */ @@ -668,7 +675,10 @@ AllocSetDelete(MemoryContext context) AllocBlock next = block->next; if (!IsKeeperBlock(set, block)) + { context->mem_allocated -= block->endptr - ((char *) block); + deallocation += block->endptr - ((char *) block); + } #ifdef CLOBBER_FREED_MEMORY wipe_mem(block, block->freeptr - ((char *) block)); @@ -681,6 +691,7 @@ AllocSetDelete(MemoryContext context) } Assert(context->mem_allocated == keepersize); + release_backend_memory(deallocation + context->mem_allocated, PG_ALLOC_ASET); /* Finally, free the context header, including the keeper block */ free(set); @@ -725,7 +736,7 @@ AllocSetAlloc(MemoryContext context, Size size) #endif blksize = chunk_size + ALLOC_BLOCKHDRSZ + ALLOC_CHUNKHDRSZ; - block = (AllocBlock) malloc(blksize); + block = (AllocBlock) malloc_backend(blksize, PG_ALLOC_ASET); if (block == NULL) return NULL; @@ -925,7 +936,7 @@ AllocSetAlloc(MemoryContext context, Size size) blksize <<= 1; /* Try to allocate it */ - block = (AllocBlock) malloc(blksize); + block = (AllocBlock) malloc_backend(blksize, PG_ALLOC_ASET); /* * We could be asking for pretty big blocks here, so cope if malloc @@ -936,7 +947,7 @@ AllocSetAlloc(MemoryContext context, Size size) blksize >>= 1; if (blksize < required_size) break; - block = (AllocBlock) malloc(blksize); + block = (AllocBlock) malloc_backend(blksize, PG_ALLOC_ASET); } if (block == NULL) @@ -1044,7 +1055,7 @@ AllocSetFree(void *pointer) #ifdef CLOBBER_FREED_MEMORY wipe_mem(block, block->freeptr - ((char *) block)); #endif - free(block); + free_backend(block, block->endptr - ((char *) block), PG_ALLOC_ASET); } else { @@ -1159,8 +1170,7 @@ AllocSetRealloc(void *pointer, Size size) /* Do the realloc */ blksize = chksize + ALLOC_BLOCKHDRSZ + ALLOC_CHUNKHDRSZ; oldblksize = block->endptr - ((char *) block); - - block = (AllocBlock) realloc(block, blksize); + block = (AllocBlock) realloc_backend(block, blksize, oldblksize, PG_ALLOC_ASET); if (block == NULL) { /* Disallow access to the chunk header. */ diff --git a/src/backend/utils/mmgr/generation.c b/src/backend/utils/mmgr/generation.c index 92401ccf73..b2bc39b472 100644 --- a/src/backend/utils/mmgr/generation.c +++ b/src/backend/utils/mmgr/generation.c @@ -37,6 +37,7 @@ #include "lib/ilist.h" #include "port/pg_bitutils.h" +#include "utils/backend_status.h" #include "utils/memdebug.h" #include "utils/memutils.h" #include "utils/memutils_memorychunk.h" @@ -203,7 +204,7 @@ GenerationContextCreate(MemoryContext parent, * Allocate the initial block. Unlike other generation.c blocks, it * starts with the context header and its block header follows that. */ - set = (GenerationContext *) malloc(allocSize); + set = (GenerationContext *) malloc_backend(allocSize, PG_ALLOC_GENERATION); if (set == NULL) { MemoryContextStats(TopMemoryContext); @@ -279,6 +280,7 @@ GenerationReset(MemoryContext context) { GenerationContext *set = (GenerationContext *) context; dlist_mutable_iter miter; + uint64 deallocation = 0; Assert(GenerationIsValid(set)); @@ -301,9 +303,14 @@ GenerationReset(MemoryContext context) if (IsKeeperBlock(set, block)) GenerationBlockMarkEmpty(block); else + { + deallocation += block->blksize; GenerationBlockFree(set, block); + } } + release_backend_memory(deallocation, PG_ALLOC_GENERATION); + /* set it so new allocations to make use of the keeper block */ set->block = KeeperBlock(set); @@ -322,10 +329,12 @@ GenerationReset(MemoryContext context) void GenerationDelete(MemoryContext context) { + /* Reset to release all releasable GenerationBlocks */ GenerationReset(context); + /* And free the context header and keeper block */ - free(context); + free_backend(context, MAXALIGN(sizeof(GenerationContext)) + context->mem_allocated, PG_ALLOC_GENERATION); } /* @@ -365,7 +374,7 @@ GenerationAlloc(MemoryContext context, Size size) { Size blksize = required_size + Generation_BLOCKHDRSZ; - block = (GenerationBlock *) malloc(blksize); + block = (GenerationBlock *) malloc_backend(blksize, PG_ALLOC_GENERATION); if (block == NULL) return NULL; @@ -467,8 +476,7 @@ GenerationAlloc(MemoryContext context, Size size) if (blksize < required_size) blksize = pg_nextpower2_size_t(required_size); - block = (GenerationBlock *) malloc(blksize); - + block = (GenerationBlock *) malloc_backend(blksize, PG_ALLOC_GENERATION); if (block == NULL) return NULL; @@ -723,9 +731,8 @@ GenerationFree(void *pointer) * list of blocks, then return it to malloc(). */ dlist_delete(&block->node); - set->header.mem_allocated -= block->blksize; - free(block); + free_backend(block, block->blksize, PG_ALLOC_GENERATION); } /* diff --git a/src/backend/utils/mmgr/slab.c b/src/backend/utils/mmgr/slab.c index 40c1d401c4..436be421fb 100644 --- a/src/backend/utils/mmgr/slab.c +++ b/src/backend/utils/mmgr/slab.c @@ -69,6 +69,7 @@ #include "postgres.h" #include "lib/ilist.h" +#include "utils/backend_status.h" #include "utils/memdebug.h" #include "utils/memutils.h" #include "utils/memutils_memorychunk.h" @@ -359,9 +360,7 @@ SlabContextCreate(MemoryContext parent, elog(ERROR, "block size %zu for slab is too small for %zu-byte chunks", blockSize, chunkSize); - - - slab = (SlabContext *) malloc(Slab_CONTEXT_HDRSZ(chunksPerBlock)); + slab = (SlabContext *) malloc_backend(Slab_CONTEXT_HDRSZ(chunksPerBlock), PG_ALLOC_SLAB); if (slab == NULL) { MemoryContextStats(TopMemoryContext); @@ -433,6 +432,7 @@ SlabReset(MemoryContext context) SlabContext *slab = (SlabContext *) context; dlist_mutable_iter miter; int i; + uint64 deallocation = 0; Assert(SlabIsValid(slab)); @@ -453,6 +453,7 @@ SlabReset(MemoryContext context) #endif free(block); context->mem_allocated -= slab->blockSize; + deallocation += slab->blockSize; } /* walk over blocklist and free the blocks */ @@ -469,9 +470,11 @@ SlabReset(MemoryContext context) #endif free(block); context->mem_allocated -= slab->blockSize; + deallocation += slab->blockSize; } } + release_backend_memory(deallocation, PG_ALLOC_SLAB); slab->curBlocklistIndex = 0; Assert(context->mem_allocated == 0); @@ -486,8 +489,13 @@ SlabDelete(MemoryContext context) { /* Reset to release all the SlabBlocks */ SlabReset(context); - /* And free the context header */ - free(context); + + /* + * Until context header allocation is included in context->mem_allocated, + * cast to slab and decrement the header allocation + */ + free_backend(context, Slab_CONTEXT_HDRSZ(((SlabContext *) context)->chunksPerBlock), + PG_ALLOC_SLAB); } /* @@ -543,8 +551,7 @@ SlabAlloc(MemoryContext context, Size size) } else { - block = (SlabBlock *) malloc(slab->blockSize); - + block = (SlabBlock *) malloc_backend(slab->blockSize, PG_ALLOC_SLAB); if (unlikely(block == NULL)) return NULL; @@ -742,7 +749,7 @@ SlabFree(void *pointer) #ifdef CLOBBER_FREED_MEMORY wipe_mem(block, slab->blockSize); #endif - free(block); + free_backend(block, slab->blockSize, PG_ALLOC_SLAB); slab->header.mem_allocated -= slab->blockSize; } diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat index 9805bc6118..7a7f9ac1e2 100644 --- a/src/include/catalog/pg_proc.dat +++ b/src/include/catalog/pg_proc.dat @@ -5420,6 +5420,23 @@ proname => 'pg_stat_get_backend_idset', prorows => '100', proretset => 't', provolatile => 's', proparallel => 'r', prorettype => 'int4', proargtypes => '', prosrc => 'pg_stat_get_backend_idset' }, +{ oid => '9890', + descr => 'statistics: memory allocation information for backends', + proname => 'pg_stat_get_memory_allocation', prorows => '100', proisstrict => 'f', + proretset => 't', provolatile => 's', proparallel => 'r', + prorettype => 'record', proargtypes => 'int4', + proallargtypes => '{int4,oid,int4,int8,int8,int8,int8,int8}', + proargmodes => '{i,o,o,o,o,o,o,o}', + proargnames => '{pid,datid,pid,allocated_bytes,aset_allocated_bytes,dsm_allocated_bytes,generation_allocated_bytes,slab_allocated_bytes}', + prosrc => 'pg_stat_get_memory_allocation' }, +{ oid => '9891', + descr => 'statistics: global memory allocation information', + proname => 'pg_stat_get_global_memory_allocation', proisstrict => 'f', + provolatile => 's', proparallel => 'r', prorettype => 'record', + proargtypes => '', proallargtypes => '{oid,int8,int8}', + proargmodes => '{o,o,o}', + proargnames => '{datid,total_bkend_mem_bytes_available,global_dsm_allocated_bytes}', + prosrc =>'pg_stat_get_global_memory_allocation' }, { oid => '2022', descr => 'statistics: information about currently active backends', proname => 'pg_stat_get_activity', prorows => '100', proisstrict => 'f', diff --git a/src/include/storage/proc.h b/src/include/storage/proc.h index ef74f32693..b3a2322328 100644 --- a/src/include/storage/proc.h +++ b/src/include/storage/proc.h @@ -404,6 +404,10 @@ typedef struct PROC_HDR int spins_per_delay; /* Buffer id of the buffer that Startup process waits for pin on, or -1 */ int startupBufferPinWaitBufId; + /* Global dsm allocations */ + pg_atomic_uint64 global_dsm_allocation; + /* Track how much memory the backends have allocated */ + pg_atomic_uint64 total_bkend_mem_bytes; } PROC_HDR; extern PGDLLIMPORT PROC_HDR *ProcGlobal; diff --git a/src/include/utils/backend_status.h b/src/include/utils/backend_status.h index d51c840daf..5ec310c769 100644 --- a/src/include/utils/backend_status.h +++ b/src/include/utils/backend_status.h @@ -10,11 +10,13 @@ #ifndef BACKEND_STATUS_H #define BACKEND_STATUS_H +#include "postgres.h" #include "datatype/timestamp.h" #include "libpq/pqcomm.h" #include "miscadmin.h" /* for BackendType */ #include "storage/backendid.h" #include "utils/backend_progress.h" +#include "utils/memtrack.h" /* ---------- @@ -170,6 +172,10 @@ typedef struct PgBackendStatus /* query identifier, optionally computed using post_parse_analyze_hook */ uint64 st_query_id; + + /* Memory allocated to this backend, both total and subtotals by type. */ + PgBackendMemoryStatus st_memory; + } PgBackendStatus; @@ -332,6 +338,8 @@ extern uint64 pgstat_get_my_query_id(void); * generate the pgstat* views. * ---------- */ +extern LocalPgBackendStatus *pgstat_fetch_stat_local_beentry(int beid); +extern PgBackendStatus *pgstat_fetch_stat_beentry(int beid); extern int pgstat_fetch_stat_numbackends(void); extern PgBackendStatus *pgstat_get_beentry_by_backend_id(BackendId beid); extern LocalPgBackendStatus *pgstat_get_local_beentry_by_backend_id(BackendId beid); diff --git a/src/include/utils/memtrack.h b/src/include/utils/memtrack.h new file mode 100644 index 0000000000..12b56cbf58 --- /dev/null +++ b/src/include/utils/memtrack.h @@ -0,0 +1,288 @@ + +#ifndef MEMTRACK_H +#define MEMTRACK_H + +/* ---------- + * Backend memory accounting functions. + * Track how much memory each backend is using + * and place a cluster-wide limit on the total amount of backend memory. + * + * The main functions are: + * init_backend_memory() + * reserve_backend_memory() + * release_backend_memory() + * exit_backend_memory() + * + * All local variables are properly initialized at startup, so init_backend_memory() + * only needs to be called after a fork() system call. + * + * The reserve/release functions implement both a "fast path" and a "slow path". + * The fast path is used for most allocations, and it only references + * local variables. The slow path is invoked periodically; it updates + * shared memory and checks for limits on total backend memory. + * + * The following local variables represent the "TRUTH" of this backend's memory allocations. + * my_memory.allocated_bytes: total amount of memory allocated by this backend. + * my_memory.allocated_bytes_by_type[type]: subtotals by allocator type. + * + * The local values are periodically reported to pgstgt. + * The following variables hold the last reported values + * reported_memory.allocated_bytes + * reported_memory.allocated_bytes_by_type[type]: + * + * The "slow path" is invoked when my_memory.allocate_bytes exceeds these bounds. + * Once invokoked, it updates the reported values and sets new bounds. + * allocation_upper_bound: update when my_memory.allocated_bytes exceeds this + * allocation_lower_bound: update when my_memory.allocated_bytes drops below this + * allocation_allowance_refill_qty amount of memory to allocate or release before updating again. + * + * These counters are the values seen by pgstat. They are a copy of reported_memory. + * proc->st_memory.allocated_bytes: last total reported to pgstat + * proc->st_memory.allocated_bytes_by_type[type]: last reported subtotals reported to pgstat + * + * Limits on total backend memory. If max_total_bkend_bytes is zero, there is no limit. + * ProcGlobal->total_bkend_mem_bytes: total amount of memory reserved by all backends, including shared memory + * max_total_bkend_bytes: maximum amount of memory allowed to be reserved by all backends. + * + * And finally, + * initial_allocation_allowance: each backend consumes this much memory simply by existing. + * ProcGlobal->global_dsm_allocated_bytes: total amount of shared memory allocated by all backends. + * ---------- + */ + +#include <unistd.h> +#include "postgres.h" +#include "port/atomics.h" + +/* + * Define a debug macro which becomes noop() when debug is disabled. + */ +#define debug(args...) (void)0 + + +/* Various types of memory allocators we are tracking. */ +typedef enum pg_allocator_type +{ + PG_ALLOC_OTHER = 0, /* Not tracked, but part of total memory */ + PG_ALLOC_ASET, /* Allocation Set */ + PG_ALLOC_DSM, /* Dynamic shared memory */ + PG_ALLOC_GENERATION, /* Generation Context (all freed at once) */ + PG_ALLOC_SLAB, /* Slab allocator */ + PG_ALLOC_TYPE_MAX, /* (Last, for array sizing) */ +} pg_allocator_type; + + +/* + * PgBackendMemoryStatus + * + * For each backend, track how much memory has been allocated. + * Note may be possible to have negative values, say if one backend + * creates DSM segments and another backend destroys them. + */ +typedef struct PgBackendMemoryStatus +{ + int64 allocated_bytes; + int64 allocated_bytes_by_type[PG_ALLOC_TYPE_MAX]; +} PgBackendMemoryStatus; + + +/* These values are candidates for GUC variables. We chose 1MV arbitrarily. */ +static const int64 initial_allocation_allowance = 1024 * 1024; /* 1MB */ +static const int64 allocation_allowance_refill_qty = 1024 * 1024; /* 1MB */ + +/* Compile time initialization constants */ +#define INIT_BACKEND_MEMORY (PgBackendMemoryStatus) \ + {initial_allocation_allowance, {initial_allocation_allowance}} +#define NO_BACKEND_MEMORY (PgBackendMemoryStatus) \ + {0, {0}} + +/* Manage memory allocation for backends. */ +extern PGDLLIMPORT PgBackendMemoryStatus my_memory; +extern PGDLLIMPORT PgBackendMemoryStatus reported_memory; +extern PGDLLIMPORT int64 allocation_upper_bound; +extern PGDLLIMPORT int64 allocation_lower_bound; + +extern PGDLLIMPORT int64 max_total_bkend_bytes; +extern PGDLLIMPORT int32 max_total_bkend_mem; + +/* These are the main entry points for backend memory accounting */ +extern void init_backend_memory(void); +static inline bool reserve_backend_memory(int64 size, pg_allocator_type type); +static inline void release_backend_memory(int64 size, pg_allocator_type type); +extern void exit_backend_memory(void); + +/* Helper functions for backend memory accounting */ +static inline bool update_local_allocation(int64 size, pg_allocator_type type); +extern bool update_global_allocation(int64 size, pg_allocator_type type); + +/* ---------- + * reserve_backend_memory() - + * Called to report a desired increase in memory for this backend. + * true if successful. + */ +static inline bool +reserve_backend_memory(int64 size, pg_allocator_type type) +{ + Assert(size >= 0); + + /* quick optimization */ + if (size == 0) + return true; + + /* CASE: the new allocation is within bounds. Take the fast path. */ + else if (my_memory.allocated_bytes + size <= allocation_upper_bound) + return update_local_allocation(size, type); + + /* CASE: out of bounds. Update pgstat and check memory limits */ + else + return update_global_allocation(size, type); +} + +/* ---------- + * unreserve_memory() - + * Called to report decrease in memory allocated for this backend. + * Note we should have already called "reserve_backend_memory" + * so we should never end up with a negative total allocation. + */ +static inline void +release_backend_memory(int64 size, pg_allocator_type type) +{ + Assert(size >= 0); + + /* quick optimization */ + if (size == 0) + return; + + /* CASE: In bounds, take the fast path */ + else if (my_memory.allocated_bytes - size >= allocation_lower_bound) + update_local_allocation(-size, type); + + /* CASE: Out of bounds. Update pgstat and memory totals */ + else + update_global_allocation(-size, type); +} + + +/* +* Fast path for reserving and releasing memory. +* This version is used for most allocations, and it +* is stripped down to the bare minimum to reduce impact +* on performance. It only updates local variables. +*/ +static inline bool +update_local_allocation(int64 size, pg_allocator_type type) +{ + /* Update our local memory counters. */ + my_memory.allocated_bytes += size; + my_memory.allocated_bytes_by_type[type] += size; + + return true; +} + + +/*-------------------------------------------- + * Convenience functions based on malloc/free + *------------------------------------------*/ + +/* + * Reserve memory from malloc if we can. + */ +static inline void * +malloc_backend(int64 size, pg_allocator_type type) +{ + void *ptr; + + /* reserve the memory if able to */ + if (!reserve_backend_memory(size, type)) + return NULL; + + /* Allocate the memory, releasing the reservation if failed */ + ptr = malloc(size); + if (ptr == NULL) + release_backend_memory(size, type); + + return ptr; +} + +/* + * Free memory which was allocated with malloc_reserved. + * Note: most mallocs have a non-portable method to + * get the size of a block of memory. Dropping the "size" parameter + * would greatly simplify the calling code. + */ +static inline void +free_backend(void *ptr, int64 size, pg_allocator_type type) +{ + release_backend_memory(size, type); + free(ptr); +} + + +/* + * Realloc reserved memory. + */ +static inline void * +realloc_backend(void *block, int64 new_size, int64 old_size, pg_allocator_type type) +{ + void *ptr; + bool success; + + /* Update the reservation to the new size */ + release_backend_memory(old_size, type); + success = reserve_backend_memory(new_size, type); + + /* If unable, free the old memory and return NULL */ + if (!success) + { + free(block); + return NULL; + } + + /* Now, actually resize the memory */ + ptr = realloc(block, new_size); + + /* + * If unable to resize, release the allocation. + * The actual memory has already been freed. + */ + if (ptr == NULL) + release_backend_memory(new_size, type); + + return ptr; +} + + +/* True if adding a and b would overflow */ +static inline bool addition_overflow(int64 a, int64 b) +{ + int64 result = a + b; + return (a > 0 && b > 0 && result < 0) || (a < 0 && b < 0 && result > 0); +} + +/* + * Helper function to add to an atomic sum, as long as the result is within bounds. + * TODO: consider moving to atomics.h + */ +static inline bool +atomic_add_within_bounds_i64(pg_atomic_uint64 *ptr, int64 add, int64 lower_bound, int64 upper_bound) +{ + int64 oldval; + int64 newval; + + for (;;) + { + oldval = (int64)pg_atomic_read_u64(ptr); + newval = oldval + add; + + /* check if we are out of bounds */ + if (newval < lower_bound || newval > upper_bound || addition_overflow(oldval, add)) + return false; + + if (pg_atomic_compare_exchange_u64(ptr, (uint64 *)&oldval, newval)) + return true; + } +} + + + +#endif //POSTGRES_IDE_MEMTRACK_H diff --git a/src/test/modules/meson.build b/src/test/modules/meson.build index 17d369e378..6bdcc6dac4 100644 --- a/src/test/modules/meson.build +++ b/src/test/modules/meson.build @@ -19,6 +19,7 @@ subdir('test_extensions') subdir('test_ginpostinglist') subdir('test_integerset') subdir('test_lfind') +subdir('test_memtrack') subdir('test_misc') subdir('test_oat_hooks') subdir('test_parser') diff --git a/src/test/modules/test_memtrack/.gitignore b/src/test/modules/test_memtrack/.gitignore new file mode 100644 index 0000000000..5dcb3ff972 --- /dev/null +++ b/src/test/modules/test_memtrack/.gitignore @@ -0,0 +1,4 @@ +# Generated subdirectories +/log/ +/results/ +/tmp_check/ diff --git a/src/test/modules/test_memtrack/Makefile b/src/test/modules/test_memtrack/Makefile new file mode 100644 index 0000000000..db44ca0575 --- /dev/null +++ b/src/test/modules/test_memtrack/Makefile @@ -0,0 +1,25 @@ +# src/test/modules/test_memtrack/Makefile + +MODULE_big = test_memtrack +OBJS = \ + $(WIN32RES) \ + test_memtrack.o \ + worker_pool.o +PGFILEDESC = "test_memtrack - example use of shared memory message queue" + +EXTENSION = test_memtrack +DATA = test_memtrack--1.0.sql + +REGRESS = test_memtrack +REGRESS_OPTS = --temp-config=$(top_srcdir)/src/test/modules/test_memtrack/test_memtrack.conf + +ifdef USE_PGXS +PG_CONFIG = pg_config +PGXS := $(shell $(PG_CONFIG) --pgxs) +include $(PGXS) +else +subdir = src/test/modules/test_memtrack +top_builddir = ../../../.. +include $(top_builddir)/src/Makefile.global +include $(top_srcdir)/contrib/contrib-global.mk +endif diff --git a/src/test/modules/test_memtrack/README b/src/test/modules/test_memtrack/README new file mode 100644 index 0000000000..75a6c1e84a --- /dev/null +++ b/src/test/modules/test_memtrack/README @@ -0,0 +1,36 @@ +test_memtrack tests the memory tracking facilities of PostgreSQL. + + +Functions +========= +test_allocation(# workers, type of memory, # blocks, block size) + + This function controls the test. It creates a pool of workers and passes them messages + asking them to allocate or free memory. It verifies the memory is allocated or freed, + and it confirms the total memory has been freed at the end of the test. + + Like test_shm_queue, this function spawns worker backends which communicate + through message queues. + + Unlike test_shm_queue, each worker is associatied with two queues, one for requests + and another for responses. All messages are exchanged between the main test backend + and the workers. The workers do not communicate with each other. + +test_memtrack(# workers, type of memory, # blocks, block size) + + This function is like test_allocation, but it only updates the memory tracking counters + and does not actually allocate or free memory. + + +Messages +======== +The message sent to the workers + request type (ALLOCATE or FREE) + type of memory + number of blocks to allocate or free + size of each block + +The response sent back. + error code (0 if no error) + starting memory usage totals and subtotals. + ending memory usage totals and subtotals. diff --git a/src/test/modules/test_memtrack/expected/test_memtrack.out b/src/test/modules/test_memtrack/expected/test_memtrack.out new file mode 100644 index 0000000000..23e51b4e80 --- /dev/null +++ b/src/test/modules/test_memtrack/expected/test_memtrack.out @@ -0,0 +1,80 @@ +CREATE EXTENSION test_memtrack; +-- Make sure we can track memory usage of a single task +-- Since logic is the same for all memory managers, only test one. +SELECT test_memtrack(1, 1, 0, 1024); + test_memtrack +--------------- + +(1 row) + +-- Make sure we can track memory usage of multiple tasks. +-- By default we are limited to 8 tasks, so stay below the limit. +SELECT test_memtrack(5, 2, 1024, 5*1024); + test_memtrack +--------------- + +(1 row) + +-- Do it again. We had a bug where the second call would fail. +SELECT test_memtrack(5, 3, 1024, 5*1024); + test_memtrack +--------------- + +(1 row) + +-- Now we're going to actualy do memory allocations. +-- We'll test each type of memory manager, except DSM. +-- Verify we can create and destroy contexts. +SELECT test_allocation(1,1,0,1024); + test_allocation +----------------- + +(1 row) + +SELECT test_allocation(1,2,0, 1024); + test_allocation +----------------- + +(1 row) + +SELECT test_allocation(1,3, 0,1024); + test_allocation +----------------- + +(1 row) + +SELECT test_allocation(1,4,0,1024); + test_allocation +----------------- + +(1 row) + +-- Create and free blocks of memory. +SELECT test_allocation(5,1,5*1024,1024); + test_allocation +----------------- + +(1 row) + +SELECT test_allocation(5,2,5,1024*1024); /* Fewer, don't exceed shmem limit */ + test_allocation +----------------- + +(1 row) + +SELECT test_allocation(5,3,5*1024,1024); + test_allocation +----------------- + +(1 row) + +SELECT test_allocation(5,4,5*1024,1024); + test_allocation +----------------- + +(1 row) + +-- Allocate more memory than we have available. +-- (this should fail because we configured max_backend_memory to 1024 Mb) +SELECT test_memtrack(5, 2, 1024, 1024*1024*1024); +ERROR: worker ran out of memory diff --git a/src/test/modules/test_memtrack/meson.build b/src/test/modules/test_memtrack/meson.build new file mode 100644 index 0000000000..4b861ab62f --- /dev/null +++ b/src/test/modules/test_memtrack/meson.build @@ -0,0 +1,38 @@ +# Copyright (c) 2022-2023, PostgreSQL Global Development Group + +test_memtrack_sources = files( + 'test_memtrack.c', + 'worker_pool.c', +) + +if host_system == 'windows' + test_memtrack_sources += rc_lib_gen.process(win32ver_rc, extra_args: [ + '--NAME', 'test_memtrack', + '--FILEDESC', 'test_memtrack - test memory tracking',]) +endif + +test_memtrack = shared_module('test_memtrack', + test_memtrack_sources, + kwargs: pg_test_mod_args, +) +test_install_libs += test_memtrack + +test_install_data += files( + 'test_memtrack.control', + 'test_memtrack--1.0.sql', +) + +tests += { + 'name': 'test_memtrack', + 'sd': meson.current_source_dir(), + 'bd': meson.current_build_dir(), + 'regress': { + 'sql': [ + 'test_memtrack', + ], + 'regress_args': [ + '--temp-config', files('test_memtrack.conf'), + ], + + }, +} diff --git a/src/test/modules/test_memtrack/sql/test_memtrack.sql b/src/test/modules/test_memtrack/sql/test_memtrack.sql new file mode 100644 index 0000000000..39318276ec --- /dev/null +++ b/src/test/modules/test_memtrack/sql/test_memtrack.sql @@ -0,0 +1,31 @@ +CREATE EXTENSION test_memtrack; + +-- Make sure we can track memory usage of a single task +-- Since logic is the same for all memory managers, only test one. +SELECT test_memtrack(1, 1, 0, 1024); + +-- Make sure we can track memory usage of multiple tasks. +-- By default we are limited to 8 tasks, so stay below the limit. +SELECT test_memtrack(5, 2, 1024, 5*1024); + +-- Do it again. We had a bug where the second call would fail. +SELECT test_memtrack(5, 3, 1024, 5*1024); + +-- Now we're going to actualy do memory allocations. +-- We'll test each type of memory manager, except DSM. + +-- Verify we can create and destroy contexts. +SELECT test_allocation(1,1,0,1024); +SELECT test_allocation(1,2,0, 1024); +SELECT test_allocation(1,3, 0,1024); +SELECT test_allocation(1,4,0,1024); + +-- Create and free blocks of memory. +SELECT test_allocation(5,1,5*1024,1024); +SELECT test_allocation(5,2,5,1024*1024); /* Fewer, don't exceed shmem limit */ +SELECT test_allocation(5,3,5*1024,1024); +SELECT test_allocation(5,4,5*1024,1024); + +-- Allocate more memory than we have available. +-- (this should fail because we configured max_backend_memory to 1024 Mb) +SELECT test_memtrack(5, 2, 1024, 1024*1024*1024); diff --git a/src/test/modules/test_memtrack/test_memtrack--1.0.sql b/src/test/modules/test_memtrack/test_memtrack--1.0.sql new file mode 100644 index 0000000000..9a39f24ee4 --- /dev/null +++ b/src/test/modules/test_memtrack/test_memtrack--1.0.sql @@ -0,0 +1,20 @@ +/* src/test/modules/test_memtrack/test_memtrack--1.0.sql */ + +-- complain if script is sourced in psql, rather than via CREATE EXTENSION +\echo Use "CREATE EXTENSION test_memtrack" to load this file. \quit + +CREATE FUNCTION test_memtrack( + num_workers pg_catalog.int4 default 1, + type pg_catalog.int4 default 1, + num_blocks pg_catalog.int4 default 1, + block_size pg_catalog.int4 default 1024) + RETURNS pg_catalog.void STRICT + AS 'MODULE_PATHNAME' LANGUAGE C; + +CREATE FUNCTION test_allocation( + num_workers pg_catalog.int4 default 1, + type pg_catalog.int4 default 1, + num_blocks pg_catalog.int4 default 1, + block_size pg_catalog.int4 default 1024) + RETURNS pg_catalog.void STRICT +AS 'MODULE_PATHNAME' LANGUAGE C; diff --git a/src/test/modules/test_memtrack/test_memtrack.c b/src/test/modules/test_memtrack/test_memtrack.c new file mode 100644 index 0000000000..e9a8a7844f --- /dev/null +++ b/src/test/modules/test_memtrack/test_memtrack.c @@ -0,0 +1,496 @@ +/*-------------------------------------------------------------------------- + * + * test_memstack.c + * Test harness code + * + * Copyright (c) 2013-2023, PostgreSQL Global Development Group + * + * IDENTIFICATION + * src/test/modules/test_memtrack/test_memtrack.c + * + * ------------------------------------------------------------------------- + */ + +#include "postgres.h" + +#include "fmgr.h" +#include "miscadmin.h" +#include "pgstat.h" +#include "varatt.h" + +#include "storage/dsm.h" +#include "utils/backend_status.h" +#include "utils/memutils_internal.h" + +#include "worker_pool.h" + +PG_MODULE_MAGIC; + +PG_FUNCTION_INFO_V1(test_memtrack); +PG_FUNCTION_INFO_V1(test_allocation); + +typedef enum {ALLOCATE, RELEASE} Action; +typedef struct AllocateRequest +{ + Action action; + pg_allocator_type type; + int32 nWorkers; + int32 nBlocks; + int64 blockSize; +} AllocateRequest; + +typedef struct ResponseData +{ + int32 errorCode; + PgBackendMemoryStatus memory; + PgBackendMemoryStatus startingMemory; +} ResponseData; + +/* Forward references */ +static bool reserveBlocks(pg_allocator_type type, int nBlocks, int blockSize); +static bool releaseBlocks(pg_allocator_type type, int nBlocks, int blockSize); +static void validateArgs(int nWorkers, pg_allocator_type type, int nBlocks, int blockSize); +static void sendRequest(WorkerPool *pool, int worker, Action action, pg_allocator_type type, int nBlocks, int blockSize); +static void processReply(WorkerPool *pool, int worker, Action actions, pg_allocator_type type, int nBlocks, int blockSize); +static Datum test_nonDSM(FunctionCallInfo fcinfo, char *workerFunction); + +/* Test the memory tracking features standalone */ +PGDLLEXPORT Datum test_memtrack(PG_FUNCTION_ARGS); +PGDLLEXPORT void test_memtrack_worker(Datum arg); + +/* Test the memory tracking features with actual allocations */ +PGDLLEXPORT Datum test_allocation(PG_FUNCTION_ARGS); +PGDLLEXPORT void test_allocation_worker(Datum arg); +/* + * Test the memory tracking features. + */ +Datum +test_memtrack(PG_FUNCTION_ARGS) +{ + return test_nonDSM(fcinfo, "test_memtrack_worker"); +} + + +Datum +test_allocation(PG_FUNCTION_ARGS) +{ + return test_nonDSM(fcinfo, "test_allocation_worker"); +} + + + +/**************** + * Parent task to test the memory tracking features. + * Schedules a pool of workers and verifies the results. + * This version tests non-shared memory allocations. + */ +static +Datum test_nonDSM(FunctionCallInfo fcinfo, char *workerFunction) +{ + WorkerPool *pool; + int64 delta, starting_bkend_bytes; + int64 expected, fudge; + + /* Get the SQL function arguments */ + int32 nWorkers = PG_GETARG_INT32(0); + pg_allocator_type type = PG_GETARG_INT32(1); + int32 nBlocks = PG_GETARG_INT32(2); + int32 blockSize = PG_GETARG_INT32(3); + + debug("nWorkers=%d type=%d nBlocks=%d blockSize=%d\n", nWorkers, type, nBlocks, blockSize); + validateArgs(nWorkers, type, nBlocks, blockSize); + + /* Our global totals may be off by an allocation allowance per worker */ + fudge = nWorkers * allocation_allowance_refill_qty * nWorkers; + expected = nWorkers * nBlocks * blockSize; + + /* Set up pool of background workers */ + pool = createWorkerPool(nWorkers, sizeof(AllocateRequest) * 10, sizeof(ResponseData) * 10, + "test_memtrack", workerFunction); + + /* Remember the total memory before we start allocations */ + starting_bkend_bytes = pg_atomic_read_u32((void *)&ProcGlobal->total_bkend_mem_bytes); + + /* Tell the workers to start their first batch of allocations */ + for (int w = 0; w < nWorkers; w++) + sendRequest(pool, w, ALLOCATE, type, nBlocks, blockSize); + + /* Get the workers response and verify all is good */ + for (int w = 0; w < nWorkers; w++) + processReply(pool, w, ALLOCATE, type, nBlocks, blockSize); + + /* Confirm the total backend memory is greater than what we just allocated */ + delta = pg_atomic_read_u32((void *)&ProcGlobal->total_bkend_mem_bytes) - starting_bkend_bytes; + debug("starting_bkend_bytes=%zd delta=%zd\n", starting_bkend_bytes, delta); + if (delta < expected - fudge || delta > expected + fudge) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("Total allocated memory %zd doesn't match expected %zd fudge=%zd\n", delta, expected, fudge))); + + /* Tell the workers to release their memory */ + for (int w=0; w < nWorkers; w++) + sendRequest(pool, w, RELEASE, type, nBlocks, blockSize); + + /* Get the workers response and verify all is good */ + for (int w = 0; w < nWorkers; w++) + processReply(pool, w, RELEASE, type, nBlocks, blockSize); + + /* Verify the new total is reasonable */ + delta = pg_atomic_read_u32((void *)&ProcGlobal->total_bkend_mem_bytes) - starting_bkend_bytes; + debug("After release: delta=%zd expected=%d\n", delta, 0); + if (delta < -fudge || delta > fudge) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("Total allocated memory %zd is less than expected %zd\n", delta, 0l))); + + /* Clean up */ + freeWorkerPool(pool); + debug("done\n"); + PG_RETURN_VOID(); +} + + +static +void validateArgs(int nWorkers, pg_allocator_type type, int nBlocks, int blockSize) +{ + /* A negative blockcount is nonsensical. */ + if (nBlocks < 0) + ereport(ERROR, + (errcode(ERRCODE_INVALID_PARAMETER_VALUE), + errmsg("repeat count size must be an integer value greater than or equal to zero"))); + + /* a minimum of 1 worker is required. */ + if (nWorkers <= 0) + ereport(ERROR, + (errcode(ERRCODE_INVALID_PARAMETER_VALUE), + errmsg("number of workers must be an integer value greater than zero"))); + + /* block size must be > 0 */ + if (blockSize <= 0) + ereport(ERROR, + (errcode(ERRCODE_INVALID_PARAMETER_VALUE), + errmsg("block size must be an integer value greater than zero"))); + + /* Valid type? */ + if (type < 0 || type >= PG_ALLOC_TYPE_MAX) + ereport(ERROR, + (errcode(ERRCODE_INVALID_PARAMETER_VALUE), + errmsg("invalid allocation type"))); +} + +static +void sendRequest(WorkerPool *pool, int worker, Action action, pg_allocator_type type, int nBlocks, int blockSize) +{ + AllocateRequest req; + int result; + debug("worker=%d action=%d type=%d nBlocks=%d blockSize=%d\n", worker, action, type, nBlocks, blockSize); + + req = (AllocateRequest) {.nBlocks = nBlocks, .action = action, .type = type, .blockSize = blockSize}; + + result = sendToWorker(pool, worker, &req, sizeof(req)); + if (result != SHM_MQ_SUCCESS) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("could not send message"))); +} + + +void processReply(WorkerPool *pool, int worker, Action action, pg_allocator_type type, int nBlocks, int blockSize) +{ + int64 delta; + ResponseData *resp; + Size len; + int result; + debug("worker=%d action=%d type=%d nBlocks=%d blockSize=%d\n", worker, action, type, nBlocks, blockSize); + + /* Receive a message. Returns a pointer to the message and a length */ + result = recvFromWorker(pool, worker, (void *)&resp, &len); + if (result != SHM_MQ_SUCCESS) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("could not receive message"))); + + if (resp->errorCode != 0) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("worker ran out of memory"))); + + /* Verify the totals */ + delta = resp->memory.allocated_bytes_by_type[type] - resp->startingMemory.allocated_bytes_by_type[type]; + debug("delta=%ld expected=%ld\n", delta, (int64)nBlocks*blockSize); + if (action == ALLOCATE && delta < nBlocks * blockSize) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("worker reported %ld bytes, expected %d", resp->memory.allocated_bytes_by_type[type], nBlocks * blockSize))); + + if (action == RELEASE && delta != 0) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("worker reported %ld bytes, expected 0", resp->memory.allocated_bytes_by_type[type]))); +} + + + +void test_memtrack_worker(Datum arg) +{ + AllocateRequest *req; + Size reqSize; + + ResponseData resp[1]; + shm_mq_result result; + + PgBackendMemoryStatus startingMemory; + + debug(""); + workerInit(arg); + + /* Now that we're running, make note of how much memory has been already allocated */ + startingMemory = my_memory; + + do + { + result = workerRecv((void *)&req, &reqSize); + debug("Received result=%d action=%d type=%d nBlocks=%d blockSize=%ld\n", result, req->action, req->type, req->nBlocks, req->blockSize); + if (result != SHM_MQ_SUCCESS) + break; + + /* Allocate or release the blocks */ + if (req->action == ALLOCATE && reserveBlocks(req->type, req->nBlocks, req->blockSize)) + resp->errorCode = 0; + else if (req->action == RELEASE && releaseBlocks(req->type, req->nBlocks, req->blockSize)) + resp->errorCode = 0; + else + resp->errorCode = 1; + + /* Get the current memory totals */ + resp->memory = my_memory; + resp->startingMemory = startingMemory; + debug("MyBEEntry=%p allocated_bytes=%ld type=%d\n", MyBEEntry, my_memory.allocated_bytes_by_type[req->type], req->type); + + /* Send the response */ + result = workerSend(resp, sizeof(resp[1])); + debug("Send errorCode=%d result=%d\n", resp->errorCode, result); + + } while (result == SHM_MQ_SUCCESS); + + workerExit(0); +} + + +bool reserveBlocks(pg_allocator_type type, int nBlocks, int blockSize) +{ + debug("type=%d nBlocks=%d blockSize=%d prev=%zd\n", type, nBlocks, blockSize, my_memory.allocated_bytes_by_type[type]); + /* Allocate the requested number of blocks */ + for (int i = 0; i < nBlocks; i++) + if (!reserve_backend_memory(blockSize, type)) + return false; + + debug("success: allocated_bytes=%zd\n", my_memory.allocated_bytes_by_type[type]); + return true; +} + + +bool releaseBlocks(pg_allocator_type type, int nBlocks, int blockSize) +{ + debug("type=%d nBlocks=%d blockSize=%d\n", type, nBlocks, blockSize); + for (int i = 0; i < nBlocks; i++) + release_backend_memory(blockSize, type); + + return true; +} + + + +/*************** + * Worker which allocates and frees blocks, + * but this one does actual allocations. + */ + +/* Forward references for the allocation worker */ +static bool freeBlocks(pg_allocator_type type, int nBlocks, int blockSize); +static MemoryContext createTestContext(pg_allocator_type type); +static bool allocateBlocks(pg_allocator_type type, int nBlocks, int blockSize); +static bool allocateDSMBlocks(int nBlocks, int blockSize); +static bool freeDSMBlocks(int nBlocks, int blockSize); +static bool allocateContextBlocks(pg_allocator_type type, int nBlocks, int blockSize); +static bool freeContextBlocks(pg_allocator_type type, int nBlocks, int blockSize); + + +/* An array of pointers to the allocated blocks */ +static void **allocations; +static MemoryContext testContext; + + +void test_allocation_worker(Datum arg) +{ + AllocateRequest *req; + Size reqSize; + + ResponseData resp[1]; + shm_mq_result result; + + PgBackendMemoryStatus startingMemory; + + debug("\n"); + workerInit(arg); + + /* Now that we're running, make note of how much memory has been already allocated */ + startingMemory = my_memory; + + do + { + result = workerRecv((void *)&req, &reqSize); + debug("Received result=%d action=%d type=%d nBlocks=%d blockSize=%ld\n", result, req->action, req->type, req->nBlocks, req->blockSize); + if (result != SHM_MQ_SUCCESS) + break; + + /* Allocate or release the blocks */ + if (req->action == ALLOCATE && allocateBlocks(req->type, req->nBlocks, req->blockSize)) + resp->errorCode = 0; + else if (req->action == RELEASE && freeBlocks(req->type, req->nBlocks, req->blockSize)) + resp->errorCode = 0; + else + resp->errorCode = 1; + + /* Get the current memory totals */ + resp->memory = my_memory; + resp->startingMemory = startingMemory; + debug("MyBEEntry=%p allocated_bytes=%ld type=%d\n", MyBEEntry, my_memory.allocated_bytes_by_type[req->type], req->type); + + /* Send the response */ + result = workerSend(resp, sizeof(resp[1])); + + } while (result == SHM_MQ_SUCCESS); + + workerExit(0); +} + + + +static bool allocateBlocks(pg_allocator_type type, int nBlocks, int blockSize) +{ + debug("type=%d nBlocks=%d blockSize=%d prev=%zd\n", type, nBlocks, blockSize, my_memory.allocated_bytes_by_type[type]); + if (type == PG_ALLOC_DSM) + return allocateDSMBlocks(nBlocks, blockSize); + else + return allocateContextBlocks(type, nBlocks, blockSize); +} + +static bool freeBlocks(pg_allocator_type type, int nBlocks, int blockSize) +{ + debug("type=%d nBlocks=%d blockSize=%d\n", type, nBlocks, blockSize); + if (type == PG_ALLOC_DSM) + return freeDSMBlocks(nBlocks, blockSize); + else + return freeContextBlocks(type, nBlocks, blockSize); +} + + + + +static +MemoryContext createTestContext(pg_allocator_type type) +{ + switch (type) + { + case PG_ALLOC_ASET: + return AllocSetContextCreate(TopMemoryContext, "test_aset", 64, 1024, 16 * 1024); + case PG_ALLOC_SLAB: + return SlabContextCreate(TopMemoryContext, "test_slab", 16 * 1024, 1024); + case PG_ALLOC_GENERATION: + return GenerationContextCreate(TopMemoryContext, "test_gen", 64, 1024, 16 * 1024); + default: return NULL; + + } +} + + +static +bool allocateContextBlocks(pg_allocator_type type, int nBlocks, int blockSize) +{ + MemoryContext old; + + debug("type=%d nBlocks=%d blockSize=%d prev=%zd\n", type, nBlocks, blockSize, my_memory.allocated_bytes_by_type[type]); + if (type == PG_ALLOC_DSM || type == PG_ALLOC_OTHER) + return false; + + /* Create a memory context for the allocations */ + testContext = createTestContext(type); + if (testContext == NULL) + return false; + + /* Switch to the new context */ + old = MemoryContextSwitchTo(testContext); + + /* Create a list of block pointers - not in the context */ + allocations = malloc(sizeof(void *) * nBlocks); + if (allocations == NULL) + return false; + + /* Allocate the requested number of blocks */ + for (int i = 0; i < nBlocks; i++) + { + allocations[i] = palloc(blockSize); + if (allocations[i] == NULL) + return false; + } + + /* Switch back to the old context */ + MemoryContextSwitchTo(old); + + debug("success: allocated_bytes=%zd\n", my_memory.allocated_bytes_by_type[type]); + return true; + } + + +static +bool freeContextBlocks(pg_allocator_type type, int nBlocks, int blockSize) +{ + debug("type=%d nBlocks=%d blockSize=%d\n", type, nBlocks, blockSize); + for (int i = 0; i < nBlocks; i++) + pfree(allocations[i]); + + MemoryContextDelete(testContext); + free(allocations); + allocations = NULL; + + return true; +} + + + + +static bool allocateDSMBlocks(int nBlocks, int blockSize) +{ + debug("nBlocks=%d blockSize=%d\n", nBlocks, blockSize); + + /* Create a list of block pointers - not in the context */ + allocations = malloc(sizeof(void *) * nBlocks); + if (allocations == NULL) + return false; + + for (int i = 0; i < nBlocks; i++) + { + allocations[i] = dsm_create(blockSize, 0); + if (allocations[i] == NULL) + return false; + debug("segment=%p size=%ld\n", allocations[i], dsm_segment_map_length(allocations[i])); + } + + return true; +} + + +static bool freeDSMBlocks(int nBlocks, int blockSize) +{ + debug("nBlocks=%d blockSize=%d\n", nBlocks, blockSize); + + for (int i = 0; i < nBlocks; i++) + dsm_detach(allocations[i]); + + free(allocations); + allocations = NULL; + + + return true; +} diff --git a/src/test/modules/test_memtrack/test_memtrack.conf b/src/test/modules/test_memtrack/test_memtrack.conf new file mode 100644 index 0000000000..256347e92c --- /dev/null +++ b/src/test/modules/test_memtrack/test_memtrack.conf @@ -0,0 +1 @@ +max_total_backend_memory = 8192 diff --git a/src/test/modules/test_memtrack/test_memtrack.control b/src/test/modules/test_memtrack/test_memtrack.control new file mode 100644 index 0000000000..548109e313 --- /dev/null +++ b/src/test/modules/test_memtrack/test_memtrack.control @@ -0,0 +1,4 @@ +comment = 'Test code for PostgreSQL memory tracking' +default_version = '1.0' +module_pathname = '$libdir/test_memtrack' +relocatable = true diff --git a/src/test/modules/test_memtrack/worker_pool.c b/src/test/modules/test_memtrack/worker_pool.c new file mode 100644 index 0000000000..8e2810a3a4 --- /dev/null +++ b/src/test/modules/test_memtrack/worker_pool.c @@ -0,0 +1,416 @@ +// +// Created by John Morris on 8/18/23. +// + + +#include "postgres.h" +#include "storage/s_lock.h" +#include "postmaster/bgworker.h" +#include "storage/dsm.h" +#include "storage/shm_toc.h" +#include "tcop/tcopprot.h" +#include "storage/spin.h" +#include "storage/procarray.h" +#include "storage/ipc.h" +#include "storage/latch.h" +#include "worker_pool.h" +#include "utils/memutils.h" +#include "miscadmin.h" +#include "utils/memtrack.h" /* for debug() */ +#include "utils/wait_event_types.h" +#include "storage/barrier.h" + + +#define WORKER_POOL_MAGIC 0x7843732 + +typedef struct WorkerPoolStartup WorkerPoolStartup; + + +struct WorkerPool +{ + dsm_segment *seg; + shm_toc *toc; + int nWorkers; + MemQue *inQ; + MemQue *outQ; + BackgroundWorkerHandle **handle; + WorkerPoolStartup *hdr; +}; + + +/* + * This structure is stored in the dynamic shared memory segment at toc(0). + * We use it to assign worker numbers to each worker. + */ +struct WorkerPoolStartup +{ + uint32 nWorkers; + pg_atomic_uint32 nextWorker; + Barrier barrier[1]; +}; + +/* Forward references */ +static int64 estimateDsmSize(int nWorkers, int inSize, int outSize); +static void cleanupWorkers(dsm_segment *seg, Datum arg); +static MemQue attachToQueue(dsm_segment *seg, shm_toc *toc, int workerIdx, int queueIdx, bool isSender); + +/* + * Create a new pool of backend workers. + */ +WorkerPool *createWorkerPool(int nWorkers, int inSize, int outSize, char *libName, char *procName) +{ + WorkerPool *pool; + int64 dsmSize; + MemoryContext oldcontext; + BackgroundWorker worker; + pid_t pid; + + debug("createWorkerPool: nWorkers=%d, inSize=%d, outSize=%d\n", nWorkers, inSize, outSize); + + /* + * We need the worker pool objects to allocated in CurTransactionContext + * rather than ExprContext; otherwise, they'll be destroyed before the on_dsm_detach + * hooks run. + */ + oldcontext = MemoryContextSwitchTo(CurTransactionContext); + + /* Create local worker pool object and allocate arrays for the headers and queues. */ + pool = MemoryContextAlloc(TopTransactionContext, sizeof(WorkerPool)); + pool->nWorkers = nWorkers; + pool->inQ = MemoryContextAlloc(TopTransactionContext, sizeof(MemQue) * nWorkers); + pool->outQ = MemoryContextAlloc(TopTransactionContext, sizeof(MemQue) * nWorkers); + pool->handle = MemoryContextAlloc(TopTransactionContext, sizeof(BackgroundWorkerHandle *) * nWorkers); + + /* Estimate the size of the shared memory and allocate shared memory */ + dsmSize = estimateDsmSize(nWorkers, inSize, outSize); + pool->seg = dsm_create(dsmSize, 0); + + /* Create table of contents in dsm so we can access contents */ + pool->toc = shm_toc_create(WORKER_POOL_MAGIC, dsm_segment_address(pool->seg), dsmSize); + + /* Set up the startup header as region 0 */ + pool->hdr = shm_toc_allocate(pool->toc, sizeof(WorkerPoolStartup)); + shm_toc_insert(pool->toc, 0, pool->hdr); + pool->hdr->nWorkers = nWorkers; + pg_atomic_init_u32(&pool->hdr->nextWorker, 0); + BarrierInit(pool->hdr->barrier, nWorkers+1); + + /* Create memory queues for each worker */ + for (int w = 0; w < nWorkers; w++) + { + shm_mq *mqIn, *mqOut; + + /* Allocate the "In" queue */ + mqIn = shm_toc_allocate(pool->toc, shm_mq_minimum_size + inSize); + shm_toc_insert(pool->toc, 1 + 2 * w, mqIn); + mqIn = shm_mq_create(mqIn, shm_mq_minimum_size + inSize); + shm_mq_set_sender(mqIn, MyProc); + + /* Allocate the "Out" queue */ + mqOut = shm_toc_allocate(pool->toc, shm_mq_minimum_size + outSize); + shm_toc_insert(pool->toc, 2 + 2 * w, mqOut); + mqOut = shm_mq_create(mqOut, shm_mq_minimum_size + outSize); + shm_mq_set_receiver(mqOut, MyProc); + } + + /* + * Arrange to kill all the workers if we abort before all workers are + * finished hooking themselves up to the dynamic shared memory segment. + * + * If we die after all the workers have finished hooking themselves up to + * the dynamic shared memory segment, we'll mark the two queues to which + * we're directly connected as detached, and the worker(s) connected to + * those queues will exit, marking any other queues to which they are + * connected as detached. This will cause any as-yet-unaware workers + * connected to those queues to exit in their turn, and so on, until + * everybody exits. + * + * But suppose the workers which are supposed to connect to the queues to + * which we're directly attached exit due to some error before theynin + * actually attach the queues. The remaining workers will have no way of + * knowing this. From their perspective, they're still waiting for those + * workers to start, when in fact they've already died. + */ + debug("configuring on_detach: seg=%p fn=%p pool=%ld\n", pool->seg, cleanupWorkers, PointerGetDatum(pool)); + on_dsm_detach(pool->seg, cleanupWorkers, PointerGetDatum(pool)); + + + /* Configure a prototypical worker. */ + worker = (BackgroundWorker) + { + .bgw_flags = BGWORKER_SHMEM_ACCESS | BGWORKER_CLASS_PARALLEL, + .bgw_start_time = BgWorkerStart_ConsistentState, + .bgw_restart_time = BGW_NEVER_RESTART, + .bgw_notify_pid = MyProcPid, + .bgw_main_arg = UInt32GetDatum(dsm_segment_handle(pool->seg)), + }; + strlcpy(worker.bgw_library_name, libName, sizeof(worker.bgw_library_name)); + strlcpy(worker.bgw_function_name, procName, sizeof(worker.bgw_function_name)); + snprintf(worker.bgw_type, sizeof(worker.bgw_type), "%s worker", libName); + snprintf(worker.bgw_name, sizeof(worker.bgw_name), "%s/%s worker for [%d]", libName, procName, MyProcPid); + debug("lib=%s proc=%s .bgw_main_arg=%d\n", libName, procName, DatumGetInt32(worker.bgw_main_arg)); + + /* Do for each worker */ + for (int w = 0; w < nWorkers; w++) + { + /* Create the worker */ + if (!RegisterDynamicBackgroundWorker(&worker, &pool->handle[w])) + ereport(ERROR, + (errcode(ERRCODE_INSUFFICIENT_RESOURCES), + errmsg("could not register background process"), + errhint("You may need to increase max_worker_processes."))); + + /* Attach the worker's memory queues */ + pool->inQ[w] = attachToQueue(pool->seg, pool->toc, w, 0, true); + pool->outQ[w] = attachToQueue(pool->seg, pool->toc, w, 1, false); + } + + /* + * Wait for workers to become ready. + * We could just wait on the barrier, but if a worker fails to reach + * the barrier, we would end up waiting forever. + */ + for (int w = 0; w < nWorkers; w++) + if (WaitForBackgroundWorkerStartup(pool->handle[w], &pid) != BGWH_STARTED) + ereport(ERROR, + (errcode(ERRCODE_INSUFFICIENT_RESOURCES), + errmsg("could not start background process"), + errhint("You may need to increase max_worker_processes."))); + + /* Wait for workers to attach to the shared memory segment */ + debug("Leader waiting on barrier\n"); + BarrierArriveAndWait(pool->hdr->barrier, 0); + + /* + * Once we reach this point, all workers are ready. We no longer need to + * kill them if we die; they'll die on their own as the message queues + * shut down. For now, kill workers on detach anyway, just to be safe. + */ + //cancel_on_dsm_detach(pool->seg, cleanupWorkers, PointerGetDatum(pool)); + + /* Resume using the original memory context */ + MemoryContextSwitchTo(oldcontext); + + return pool; +} + + +shm_mq_result sendToWorker(WorkerPool *pool, int workerIdx, void *msg, Size len) +{ + shm_mq_result result; + debug("workerIdx=%d, len=%zd\n", workerIdx, len); + + result = shm_mq_send(pool->inQ[workerIdx], len, msg, true, true); + + return result; +} + +shm_mq_result recvFromWorker(WorkerPool *pool, int workerIdx, void**msg, Size *len) +{ + shm_mq_result result; + + debug("waiting workerIdx=%d\n", workerIdx); + result = shm_mq_receive(pool->outQ[workerIdx], len, msg, false); + debug("received result=%d workerIdx=%d, len=%zd\n", result, workerIdx, *len); + + return result; +} + +void freeWorkerPool(WorkerPool *pool) +{ + /* Only free the pool once. (possibly reentrant) */ + dsm_segment *seg = pool->seg; + pool->seg = NULL; + debug("pool=%p seg=%p\n", pool, seg); + if (seg == NULL) + return; + + /* Terminate the background workers */ + cleanupWorkers(pool->seg, PointerGetDatum(pool)); + + /* Detach from the message queues */ + for (int w = 0; w < pool->nWorkers; w++) + { + shm_mq_detach(pool->inQ[w]); + shm_mq_detach(pool->outQ[w]); + } + + /* Detach and destroy the shared memory segment (if we haven't already) */ + dsm_detach(seg); + + /* Free the pool object */ + pfree(pool->inQ); + pfree(pool->outQ); + pfree(pool->handle); + pfree(pool); +} + + +/* Wrapper to remove workers when detaching from dsm */ +static void cleanupWorkers(dsm_segment *seg, Datum arg) +{ + WorkerPool *pool = (WorkerPool *)DatumGetPointer(arg); + debug("seg=%p pool=%p\n", seg, pool); + for (int w = 0; w < pool->nWorkers; w++) + TerminateBackgroundWorker(pool->handle[w]); + for (int w = 0; w < pool->nWorkers; w++) + WaitForBackgroundWorkerShutdown(pool->handle[w]); +} + + +/* + * Estimate how much shared memory we need for thw pool of workers. + * + * Because the TOC machinery may choose to insert padding of oddly-sized + * requests, we must estimate each chunk separately. + * + * We need one key to register the location of the header, and we need + * nworkers * 2 keys to track the locations of the message queues. + */ +static +int64 estimateDsmSize(int nWorkers, int inSize, int outSize) +{ + shm_toc_estimator e[1]; + + shm_toc_initialize_estimator(e); + + shm_toc_estimate_keys(e, 1 + 2 * nWorkers); + + shm_toc_estimate_chunk(e, sizeof(WorkerPoolStartup)); + + for (int w = 0; w < nWorkers; w++) + { + shm_toc_estimate_chunk(e, shm_mq_minimum_size + inSize); + shm_toc_estimate_chunk(e, shm_mq_minimum_size + outSize); + } + + return shm_toc_estimate(e); +} + + + +/* + * ---------------------------------------------------------------- + * Worker process code. + * ---------------------------------------------------------------- + */ + +static dsm_segment *seg; +static shm_toc *toc; + +static int myWorkerNumber; +static MemQue inQ; +static MemQue outQ; +static WorkerPoolStartup *hdr; + +shm_mq_result workerRecv(void **msg, Size *msgSize) +{ + debug("MyWorkerNumber=%d\n", myWorkerNumber); + return shm_mq_receive(inQ, msgSize, msg, false); +} + +shm_mq_result workerSend(void *msg, Size msgSize) +{ + debug("MyWorkerNumber=%d msgSize=%zd\n", myWorkerNumber, msgSize); + return shm_mq_send(outQ, msgSize, msg, false, true); +} + + +/* + * This must be called from the worker process's main function. + */ +void workerInit(Datum arg) +{ + dsm_handle handle; + + /* We are passed the dsm handle of the worker pool */ + handle = DatumGetInt32(arg); + debug("handle=%d\n", handle); + + /* + * Establish signal handlers. + * + * We want CHECK_FOR_INTERRUPTS() to kill off this worker process just as + * it would a normal user backend. To make that happen, we use die(). + */ + pqsignal(SIGTERM, die); + BackgroundWorkerUnblockSignals(); + + /* + * Connect to the dynamic shared memory segment. + * + * The backend that registered this worker passed us the ID of a shared + * memory segment to which we must attach for further instructions. Once + * we've mapped the segment in our address space, attach to the table of + * contents so we can locate the various data structures we'll need to + * find within the segment. + * + * Note: at this point, we have not created any ResourceOwner in this + * process. This will result in our DSM mapping surviving until process + * exit, which is fine. If there were a ResourceOwner, it would acquire + * ownership of the mapping, but we have no need for that. + */ + seg = dsm_attach(handle); + if (seg == NULL) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("unable to map dynamic shared memory segment"))); + toc = shm_toc_attach(WORKER_POOL_MAGIC, dsm_segment_address(seg)); + if (toc == NULL) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("bad magic number in dynamic shared memory segment"))); + + /* Attach to the startup header and get our worker idx */ + hdr = shm_toc_lookup(toc, 0, false); + myWorkerNumber = pg_atomic_fetch_add_u32(&hdr->nextWorker, 1); + debug("myWorkerNumber=%d\n", myWorkerNumber); + if (myWorkerNumber >= hdr->nWorkers) + ereport(ERROR, + (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE), + errmsg("too many workers created in pool. Max=%d", hdr->nWorkers))); + + + /* Attach to the in and out message queues. */ + inQ = attachToQueue(seg, toc, myWorkerNumber, 0, false); + outQ = attachToQueue(seg, toc, myWorkerNumber, 1, true); + + /* Wait for everybody else to become ready */ + debug("Waiting on barrier\n"); + BarrierArriveAndWait(hdr->barrier, 0); + debug("Barrier passed\n"); +} + + +void workerExit(int code) +{ + debug("code=%d\n", code); + /* + * We're done. For cleanliness, explicitly detach from the shared memory + * segment (that would happen anyway during process exit, though). + */ + //dsm_detach(seg); + proc_exit(code); +} + + +static MemQue attachToQueue(dsm_segment *seg, shm_toc *toc, int workerIdx, int queueIdx, bool isSender) +{ + MemQue que; + shm_mq *mq; + + /* Attach to the appropriate message queues. */ + mq = shm_toc_lookup(toc, 1 + 2 * workerIdx + queueIdx, false); + + /* Make note whether we are sending or receiving */ + if (isSender) + shm_mq_set_sender(mq, MyProc); + else + shm_mq_set_receiver(mq, MyProc); + + /* Attach to the queue */ + que = shm_mq_attach(mq, seg, NULL); + + debug("workerIdx=%d, queueIdx=%d handle=%p\n", workerIdx, queueIdx, que); + return que; +} diff --git a/src/test/modules/test_memtrack/worker_pool.h b/src/test/modules/test_memtrack/worker_pool.h new file mode 100644 index 0000000000..b4c2a84939 --- /dev/null +++ b/src/test/modules/test_memtrack/worker_pool.h @@ -0,0 +1,69 @@ +/* + * Implement a pool of workers that can be used to perform work in parallel. + * This version is derived from the postgres test_shm_mq module, but it is + * intended to be more general purpose. Besides being useful for running + * tests, it could eventually replace the current implementation of parallel query. + * For now, it is just a piece of infrastructure for running parallel tests. + * + * The worker pool is created by the owner process. It creates a shared memory + * segment and a set of background workers. Each background worker is provided + * with three message queues: + * - inQ: used to send messages to the worker + * - outQ: used to reply to the owner + * - errQ: used to send error exceptions back to the owner. + * Note: errQ is not yet implemented. + * + * The flow of control for the owner is: + * pool = createWorkerPool(nWorkers, inSize, outSize, libName, procName) + * repeat + * sendToWorker(pool, workerIdx, msg, len) + * recvFromWorker(pool, workerIdx, &msg, &len) + * until done + * freeWorkerPool(pool) + * + * The flow of control for the worker is: + * workerInit(arg) + * repeat + * workerRecv(&msg, &len) + * workerSend(msg, len) + * until done + * workerExit() + * + * If the workers don't exit on their own, they will be terminated when + * the owner calls freeWorkerPool. + * + * Currently, the worker entry point is passed as the name of a procedure + * and the name of a shared library. + * Future updates may allow the worker to be specified as function pointer + * in the main postgres executable. + */ + +#ifndef WORKER_POOL_H +#define WORKER_POOL_H +#include <unistd.h> +#include "postgres.h" +#include "storage/dsm.h" +#include "storage/shm_toc.h" +#include "storage/shm_mq.h" +#include "storage/s_lock.h" + +typedef struct WorkerPool WorkerPool; +typedef shm_mq_handle *MemQue; + +/* + * Procedures called from the worker pool owner + */ +WorkerPool *createWorkerPool(int nWorkers, int inSize, int outSize, char *libName, char *procName); +shm_mq_result sendToWorker(WorkerPool *pool, int workerIdx, void *msg, Size len); +shm_mq_result recvFromWorker(WorkerPool *pool, int workerIdx, void **msg, Size *len); +void freeWorkerPool(WorkerPool *pool); + +/* + * Procedures called from the worker + */ +void workerInit(Datum arg); +shm_mq_result workerRecv(void **msg, Size *msgSize); +shm_mq_result workerSend(void *msg, Size msgSize); +void workerExit(int code); + +#endif //WORKER_POOL_H diff --git a/src/test/regress/expected/rules.out b/src/test/regress/expected/rules.out index 5058be5411..171dffa456 100644 --- a/src/test/regress/expected/rules.out +++ b/src/test/regress/expected/rules.out @@ -1873,6 +1873,26 @@ pg_stat_database_conflicts| SELECT oid AS datid, pg_stat_get_db_conflict_startup_deadlock(oid) AS confl_deadlock, pg_stat_get_db_conflict_logicalslot(oid) AS confl_active_logicalslot FROM pg_database d; +pg_stat_global_memory_allocation| WITH sums AS ( + SELECT sum(pg_stat_memory_allocation.aset_allocated_bytes) AS total_aset_allocated_bytes, + sum(pg_stat_memory_allocation.dsm_allocated_bytes) AS total_dsm_allocated_bytes, + sum(pg_stat_memory_allocation.generation_allocated_bytes) AS total_generation_allocated_bytes, + sum(pg_stat_memory_allocation.slab_allocated_bytes) AS total_slab_allocated_bytes + FROM pg_stat_memory_allocation + ) + SELECT s.datid, + current_setting('shared_memory_size'::text, true) AS shared_memory_size, + (current_setting('shared_memory_size_in_huge_pages'::text, true))::integer AS shared_memory_size_in_huge_pages, + pg_size_bytes(current_setting('max_total_backend_memory'::text, true)) AS max_total_backend_memory_bytes, + s.total_bkend_mem_bytes_available, + s.global_dsm_allocated_bytes, + sums.total_aset_allocated_bytes, + sums.total_dsm_allocated_bytes, + sums.total_generation_allocated_bytes, + sums.total_slab_allocated_bytes + FROM sums, + (pg_stat_get_global_memory_allocation() s(datid, total_bkend_mem_bytes_available, global_dsm_allocated_bytes) + LEFT JOIN pg_database d ON ((s.datid = d.oid))); pg_stat_gssapi| SELECT pid, gss_auth AS gss_authenticated, gss_princ AS principal, @@ -1899,6 +1919,15 @@ pg_stat_io| SELECT backend_type, fsync_time, stats_reset FROM pg_stat_get_io() b(backend_type, object, context, reads, read_time, writes, write_time, writebacks, writeback_time, extends, extend_time, op_bytes, hits, evictions, reuses, fsyncs, fsync_time, stats_reset); +pg_stat_memory_allocation| SELECT s.datid, + s.pid, + s.allocated_bytes, + s.aset_allocated_bytes, + s.dsm_allocated_bytes, + s.generation_allocated_bytes, + s.slab_allocated_bytes + FROM (pg_stat_get_memory_allocation(NULL::integer) s(datid, pid, allocated_bytes, aset_allocated_bytes, dsm_allocated_bytes, generation_allocated_bytes, slab_allocated_bytes) + LEFT JOIN pg_database d ON ((s.datid = d.oid))); pg_stat_progress_analyze| SELECT s.pid, s.datid, d.datname, diff --git a/src/test/regress/expected/stats.out b/src/test/regress/expected/stats.out index 94187e59cf..8e70c836cd 100644 --- a/src/test/regress/expected/stats.out +++ b/src/test/regress/expected/stats.out @@ -1628,4 +1628,40 @@ SELECT COUNT(*) FROM brin_hot_3 WHERE a = 2; DROP TABLE brin_hot_3; SET enable_seqscan = on; +-- ensure that allocated_bytes exist for backends +SELECT + allocated_bytes >= 0 AS result +FROM + pg_stat_activity ps + JOIN pg_stat_memory_allocation pa ON (pa.pid = ps.pid) +WHERE + backend_type IN ('checkpointer', 'background writer', 'walwriter', 'autovacuum launcher'); + result +-------- + t + t + t + t +(4 rows) + +-- ensure that pg_stat_global_memory_allocation view exists +SELECT + datid > 0, pg_size_bytes(shared_memory_size) >= 0, shared_memory_size_in_huge_pages >= -1, global_dsm_allocated_bytes >= 0 +FROM + pg_stat_global_memory_allocation; + ?column? | ?column? | ?column? | ?column? +----------+----------+----------+---------- + t | t | t | t +(1 row) + +-- ensure that pg_stat_memory_allocation view exists +SELECT + pid > 0, allocated_bytes >= 0, aset_allocated_bytes >= 0, dsm_allocated_bytes >= 0, generation_allocated_bytes >= 0, slab_allocated_bytes >= 0 +FROM + pg_stat_memory_allocation limit 1; + ?column? | ?column? | ?column? | ?column? | ?column? | ?column? +----------+----------+----------+----------+----------+---------- + t | t | t | t | t | t +(1 row) + -- End of Stats Test diff --git a/src/test/regress/sql/stats.sql b/src/test/regress/sql/stats.sql index 1e21e55c6d..cdc8508309 100644 --- a/src/test/regress/sql/stats.sql +++ b/src/test/regress/sql/stats.sql @@ -840,4 +840,24 @@ DROP TABLE brin_hot_3; SET enable_seqscan = on; +-- ensure that allocated_bytes exist for backends +SELECT + allocated_bytes >= 0 AS result +FROM + pg_stat_activity ps + JOIN pg_stat_memory_allocation pa ON (pa.pid = ps.pid) +WHERE + backend_type IN ('checkpointer', 'background writer', 'walwriter', 'autovacuum launcher'); + +-- ensure that pg_stat_global_memory_allocation view exists +SELECT + datid > 0, pg_size_bytes(shared_memory_size) >= 0, shared_memory_size_in_huge_pages >= -1, global_dsm_allocated_bytes >= 0 +FROM + pg_stat_global_memory_allocation; + +-- ensure that pg_stat_memory_allocation view exists +SELECT + pid > 0, allocated_bytes >= 0, aset_allocated_bytes >= 0, dsm_allocated_bytes >= 0, generation_allocated_bytes >= 0, slab_allocated_bytes >= 0 +FROM + pg_stat_memory_allocation limit 1; -- End of Stats Test -- 2.33.0 ^ permalink raw reply [nested|flat] 279+ messages in thread
end of thread, other threads:[~2023-08-31 16:18 UTC | newest] Thread overview: 279+ messages (download: mbox mbox.gz follow: Atom feed) -- links below jump to the message on this page -- 2018-11-07 07:53 [PATCH 3/5] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2018-11-07 07:53 [PATCH v24 3/5] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v33 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v44 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v40 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v37 3/6] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v33 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v42 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v36 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v30 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v25 4/8] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v34 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v38 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v35 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v43 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v41 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v49 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v47 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v46 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v31 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v29 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v45 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v39 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v32 4/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v52 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v48 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v50 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2020-03-13 07:59 [PATCH v51 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v53 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v54 3/7] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2021-03-11 09:01 [PATCH v55 3/3] Make archiver process an auxiliary process Kyotaro Horiguchi <[email protected]> 2023-01-05 18:58 Re: Add tracking of backend memory allocated to pg_stat_activity Reid Thompson <[email protected]> 2023-01-05 20:13 ` Re: Add tracking of backend memory allocated to pg_stat_activity Justin Pryzby <[email protected]> 2023-01-05 21:33 ` Re: Add tracking of backend memory allocated to pg_stat_activity Reid Thompson <[email protected]> 2023-08-31 16:18 ` Re: Add tracking of backend memory allocated to pg_stat_activity John Morris <[email protected]>
This inbox is served by agora; see mirroring instructions for how to clone and mirror all data and code used for this inbox