public inbox for [email protected]
help / color / mirror / Atom feedFrom: Hayato Kuroda (Fujitsu) <[email protected]>
To: 'Amit Kapila' <[email protected]>
Cc: [email protected] <[email protected]>
Cc: Michael Paquier <[email protected]>
Cc: David G. Johnston <[email protected]>
Cc: Bertrand Drouvot <[email protected]>
Cc: Zhijie Hou (Fujitsu) <[email protected]>
Subject: RE: ReplicationSlotRelease() crashes when the instance is in the single user mode
Date: Thu, 27 Feb 2025 07:58:40 +0000
Message-ID: <OSCPR01MB149669CFD736D48C4450337DDF5CD2@OSCPR01MB14966.jpnprd01.prod.outlook.com> (raw)
In-Reply-To: <CAA4eK1LK5nZUWwNY31X6BUXdQwwzpKGKxnw+zFf-cWKm3nuCEg@mail.gmail.com>
References: <OSCPR01MB14966ED588A0328DAEBE8CB25F5FA2@OSCPR01MB14966.jpnprd01.prod.outlook.com>
<[email protected]>
<OSCPR01MB149669EF7D087A588773B7E04F5FA2@OSCPR01MB14966.jpnprd01.prod.outlook.com>
<[email protected]>
<OSCPR01MB14966559522B9699E1F769128F5FA2@OSCPR01MB14966.jpnprd01.prod.outlook.com>
<OSCPR01MB14966F234CD3121F51810D5CCF5C52@OSCPR01MB14966.jpnprd01.prod.outlook.com>
<[email protected]>
<OSCPR01MB14966FCE16ECCF29AE0502AF2F5C42@OSCPR01MB14966.jpnprd01.prod.outlook.com>
<OSCPR01MB149660520A0E53CC53E8B66F1F5C42@OSCPR01MB14966.jpnprd01.prod.outlook.com>
<CAA4eK1LK5nZUWwNY31X6BUXdQwwzpKGKxnw+zFf-cWKm3nuCEg@mail.gmail.com>
Dear Amit,
> Shouldn't such a check be present in the CheckSlotPermissions() kind
> of function to perform it in the central place?
OK. I checked whether we can reuse pre-existing functions, but it seems not appropriate.
CheckSlotPermissions() is called even by pg_drop_replication_slot(), and
CheckSlotRequirements() is not called by pg_sync_replication_slots(). I defined new
function CheckSlotIsInSingleUserMode() and put the check there.
I removed function name from the ereport(), but I feel it is enough - CheckSlotPermissions()
and CheckSlotRequirements() do not have.
> > For now, functions for replication origin and replication messages were
> retained.
> > I can handle them after the discussion.
> >
>
> Which other functions do we see similar restrictions? I checked
> "sequence manipulation functions" (1), and "Transaction ID and
> Snapshot Information Functions" (2) but couldn't see similar
> restrictions.
>
> (1) - https://www.postgresql.org/docs/current/functions-sequence.html
> (2) -
> https://www.postgresql.org/docs/current/functions-info.html#FUNCTIONS-INF
> O-SNAPSHOT
I grepped sources and could not find explicit limitations neither. So...this might
be overkill.　But I still think the restriction is OK for the slot - no need to do
some efforts for accepting single-user mode, just close the cover.
Best regards,
Hayato Kuroda
FUJITSU LIMITED
Attachments:
[application/octet-stream] v4-0001-Prohibit-slot-operations-while-in-the-single-user.patch (3.9K, ../OSCPR01MB149669CFD736D48C4450337DDF5CD2@OSCPR01MB14966.jpnprd01.prod.outlook.com/2-v4-0001-Prohibit-slot-operations-while-in-the-single-user.patch)
download | inline diff:
From b19815119909d0529d6c52889ab0959ec44e7e7f Mon Sep 17 00:00:00 2001
From: Hayato Kuroda <[email protected]>
Date: Wed, 19 Feb 2025 11:37:26 +0900
Subject: [PATCH v4] Prohibit slot operations while in the single user mode
---
src/backend/replication/logical/logicalfuncs.c | 2 ++
src/backend/replication/slot.c | 12 ++++++++++++
src/backend/replication/slotfuncs.c | 12 ++++++++++++
src/include/replication/slot.h | 1 +
4 files changed, 27 insertions(+)
diff --git a/src/backend/replication/logical/logicalfuncs.c b/src/backend/replication/logical/logicalfuncs.c
index ca53caac2f..178a07af7d 100644
--- a/src/backend/replication/logical/logicalfuncs.c
+++ b/src/backend/replication/logical/logicalfuncs.c
@@ -113,6 +113,8 @@ pg_logical_slot_get_changes_guts(FunctionCallInfo fcinfo, bool confirm, bool bin
List *options = NIL;
DecodingOutputState *p;
+ CheckSlotIsInSingleUserMode();
+
CheckSlotPermissions();
CheckLogicalDecodingRequirements();
diff --git a/src/backend/replication/slot.c b/src/backend/replication/slot.c
index 719e531eb9..d041adfe48 100644
--- a/src/backend/replication/slot.c
+++ b/src/backend/replication/slot.c
@@ -1442,6 +1442,18 @@ CheckSlotPermissions(void)
"REPLICATION")));
}
+/*
+ * Check whether the instance is in single-user mode.
+ */
+void
+CheckSlotIsInSingleUserMode(void)
+{
+ if (!IsUnderPostmaster)
+ ereport(ERROR,
+ (errcode(ERRCODE_OBJECT_NOT_IN_PREREQUISITE_STATE),
+ errmsg("replication slots cannot be used in single-user mode")));
+}
+
/*
* Reserve WAL for the currently active slot.
*
diff --git a/src/backend/replication/slotfuncs.c b/src/backend/replication/slotfuncs.c
index 146eef5871..b669774c87 100644
--- a/src/backend/replication/slotfuncs.c
+++ b/src/backend/replication/slotfuncs.c
@@ -17,6 +17,7 @@
#include "access/xlogrecovery.h"
#include "access/xlogutils.h"
#include "funcapi.h"
+#include "miscadmin.h"
#include "replication/logical.h"
#include "replication/slot.h"
#include "replication/slotsync.h"
@@ -76,6 +77,8 @@ pg_create_physical_replication_slot(PG_FUNCTION_ARGS)
if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
+ CheckSlotIsInSingleUserMode();
+
CheckSlotPermissions();
CheckSlotRequirements();
@@ -182,6 +185,8 @@ pg_create_logical_replication_slot(PG_FUNCTION_ARGS)
if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
+ CheckSlotIsInSingleUserMode();
+
CheckSlotPermissions();
CheckLogicalDecodingRequirements();
@@ -515,6 +520,8 @@ pg_replication_slot_advance(PG_FUNCTION_ARGS)
Assert(!MyReplicationSlot);
+ CheckSlotIsInSingleUserMode();
+
CheckSlotPermissions();
if (XLogRecPtrIsInvalid(moveto))
@@ -612,9 +619,12 @@ copy_replication_slot(FunctionCallInfo fcinfo, bool logical_slot)
TupleDesc tupdesc;
HeapTuple tuple;
+
if (get_call_result_type(fcinfo, NULL, &tupdesc) != TYPEFUNC_COMPOSITE)
elog(ERROR, "return type must be a row type");
+ CheckSlotIsInSingleUserMode();
+
CheckSlotPermissions();
if (logical_slot)
@@ -871,6 +881,8 @@ pg_sync_replication_slots(PG_FUNCTION_ARGS)
char *err;
StringInfoData app_name;
+ CheckSlotIsInSingleUserMode();
+
CheckSlotPermissions();
if (!RecoveryInProgress())
diff --git a/src/include/replication/slot.h b/src/include/replication/slot.h
index f5a24ccfbf..24523b1001 100644
--- a/src/include/replication/slot.h
+++ b/src/include/replication/slot.h
@@ -306,6 +306,7 @@ extern void CheckPointReplicationSlots(bool is_shutdown);
extern void CheckSlotRequirements(void);
extern void CheckSlotPermissions(void);
+extern void CheckSlotIsInSingleUserMode(void);
extern ReplicationSlotInvalidationCause
GetSlotInvalidationCause(const char *invalidation_reason);
extern const char *GetSlotInvalidationCauseName(ReplicationSlotInvalidationCause cause);
--
2.43.5
view thread (31+ messages) latest in thread
reply
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: [email protected]
Cc: [email protected], [email protected], [email protected], [email protected], [email protected], [email protected], [email protected]
Subject: RE: ReplicationSlotRelease() crashes when the instance is in the single user mode
In-Reply-To: <OSCPR01MB149669CFD736D48C4450337DDF5CD2@OSCPR01MB14966.jpnprd01.prod.outlook.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox