pg.ddx.io  pgaudit/pgaudit GitHub issues and pull requests  
help / color / mirror / Atom feed
[pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
10+ messages / 3 participants
[nested] [flat]

* [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-09-30 18:10  "zjyzh (@zjyzh)" <noreply+zjyzh@github.com>
  0 siblings, 0 replies; 10+ messages in thread

From: zjyzh (@zjyzh) @ 2026-09-30 18:10 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

## Description of the change

Fixes #333.

### Bug
On PostgreSQL 19, the pgaudit regression suite crashes the backend in the FK trigger test when `pgaudit.log_rows` is on, and FK checks produce extra audit lines when it is off.
```
LOG:  client backend (PID 8949) was terminated by signal 11: Segmentation fault
not ok 1     - pgaudit
```
```
+NOTICE:  AUDIT: OBJECT,67,1,READ,SELECT,TABLE,public.aaa,INSERT INTO bbb VALUES (1),<not logged>
+NOTICE:  AUDIT: SESSION,67,1,READ,SELECT,TABLE,public.aaa,INSERT INTO bbb VALUES (1),<not logged>
```

### Root cause
PostgreSQL commit [`5aeb36f3273`](https://github.com/postgres/postgres/commit/5aeb36f3273c768831284082f85b63066e84f16d) makes the RI fast path call `ExecutorCheckPerms_hook` from inside the FK trigger, with lists built in a per-tuple context that is reset after each check. pgaudit treated every call as the top stack item's own check. It saved those pointers and read them after they were freed. It also failed deferred FK checks at COMMIT with `pgaudit stack is empty`, and it logged FK reads against unrelated statements.

### Fix
A new `permsPending` flag on the audit event marks the check a statement owns: the first check made during its `ExecutorStart`, or COPY's. Only that check is audited, so RI fast-path checks are skipped, as they were before `5aeb36f3273`. The change is `+19/-1` in `pgaudit.c`, plus regression cases for immediate and deferred FKs, `SET CONSTRAINTS`, and `COPY`, with `pgaudit.log_rows` off and on.

## Branches
`integration`. Only PostgreSQL 19 is affected, because the RI fast path does not exist in 18 and earlier.

## Testing
`make check` in `contrib/pgaudit` against PostgreSQL `REL_19_STABLE`, which includes `5aeb36f3273`: `ok 1 - pgaudit`. Without the fix, the same suite crashes with signal 11.

inline patch 1: expected/pgaudit.out, pgaudit.c, sql/pgaudit.sql (+235/-10 lines) download inline-1.patchdiff --git a/expected/pgaudit.out b/expected/pgaudit.out index f27c370..ad5ae4a 100644 --- a/expected/pgaudit.out +++ b/expected/pgaudit.out @@ -1460,6 +1460,64 @@ NOTICE: AUDIT: SESSION,107,1,DDL,DROP TABLE,,,DROP TABLE fk_fk,<none> DROP TABLE fk_pk; NOTICE: AUDIT: SESSION,108,1,DDL,DROP TABLE,,,DROP TABLE fk_pk,<none> -- +-- Test that FK checks made by the RI fast path are not logged against the +-- statement. Since PostgreSQL 19 the fast path checks permissions from the FK +-- trigger, which runs after the statement's own check or, when the constraint +-- is deferred, at COMMIT. A top-level COMMIT runs with the stack empty, while +-- one made by a procedure or DO block runs with the CALL or DO on top. +CREATE TABLE ri_pk (id int PRIMARY KEY); +NOTICE: AUDIT: SESSION,109,1,DDL,CREATE TABLE,,,CREATE TABLE ri_pk (id int PRIMARY KEY),<none> +CREATE TABLE ri_fk (id int REFERENCES ri_pk (id) DEFERRABLE); +NOTICE: AUDIT: SESSION,110,1,DDL,CREATE TABLE,,,CREATE TABLE ri_fk (id int REFERENCES ri_pk (id) DEFERRABLE),<none> +INSERT INTO ri_pk VALUES (1), (2); +NOTICE: AUDIT: SESSION,111,1,WRITE,INSERT,TABLE,public.ri_pk,"INSERT INTO ri_pk VALUES (1), (2)",<none> +INSERT INTO ri_fk VALUES (1); +NOTICE: AUDIT: SESSION,112,1,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (1),<none> +BEGIN; +SET CONSTRAINTS ALL DEFERRED; +INSERT INTO ri_fk VALUES (2); +NOTICE: AUDIT: SESSION,113,1,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (2),<none> +COMMIT; +COPY ri_fk FROM stdin; +NOTICE: AUDIT: SESSION,114,1,WRITE,COPY,TABLE,public.ri_fk,COPY ri_fk FROM stdin,<none> +CREATE PROCEDURE ri_proc() LANGUAGE plpgsql AS $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; + INSERT INTO ri_fk VALUES (2); +END $$; +NOTICE: AUDIT: SESSION,115,1,DDL,CREATE PROCEDURE,,,"CREATE PROCEDURE ri_proc() LANGUAGE plpgsql AS $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; + INSERT INTO ri_fk VALUES (2); +END $$",<none> +CALL ri_proc(); +NOTICE: AUDIT: SESSION,116,1,FUNCTION,EXECUTE,FUNCTION,public.ri_proc,CALL ri_proc(),<none> +NOTICE: AUDIT: SESSION,116,2,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (1),<none> +NOTICE: AUDIT: SESSION,116,3,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (2),<none> +DO $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; +END $$; +NOTICE: AUDIT: SESSION,117,1,FUNCTION,DO,,,"DO $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; +END $$",<none> +NOTICE: AUDIT: SESSION,117,2,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (1),<none> +DROP PROCEDURE ri_proc(); +NOTICE: AUDIT: SESSION,118,1,DDL,DROP PROCEDURE,,,DROP PROCEDURE ri_proc(),<none> +DROP TABLE ri_fk; +NOTICE: AUDIT: SESSION,119,1,DDL,DROP TABLE,,,DROP TABLE ri_fk,<none> +DROP TABLE ri_pk; +NOTICE: AUDIT: SESSION,120,1,DDL,DROP TABLE,,,DROP TABLE ri_pk,<none> +-- -- Test rows retrived or affected by statements \connect - :current_user SET pgaudit.log = 'all'; @@ -2650,6 +2708,65 @@ NOTICE: AUDIT: SESSION,150,1,DDL,DROP INDEX,,,DROP INDEX h_idx,<none>,0 DROP TABLE h; NOTICE: AUDIT: SESSION,151,1,DDL,DROP TABLE,,,DROP TABLE h,<none>,0 -- +-- Test that FK checks made by the RI fast path are not logged against the +-- statement. Since PostgreSQL 19 the fast path checks permissions from the FK +-- trigger, which runs after the statement's own check or, when the constraint +-- is deferred, at SET CONSTRAINTS ... IMMEDIATE or COMMIT, including a COMMIT +-- made by a procedure or DO block. +CREATE TABLE ri_pk (id int PRIMARY KEY); +NOTICE: AUDIT: SESSION,152,1,DDL,CREATE TABLE,,,CREATE TABLE ri_pk (id int PRIMARY KEY),<none>,0 +CREATE TABLE ri_fk (id int REFERENCES ri_pk (id) DEFERRABLE); +NOTICE: AUDIT: SESSION,153,1,DDL,CREATE TABLE,,,CREATE TABLE ri_fk (id int REFERENCES ri_pk (id) DEFERRABLE),<none>,0 +INSERT INTO ri_pk VALUES (1), (2); +NOTICE: AUDIT: SESSION,154,1,WRITE,INSERT,TABLE,public.ri_pk,"INSERT INTO ri_pk VALUES (1), (2)",<none>,2 +INSERT INTO ri_fk VALUES (1), (2); +NOTICE: AUDIT: SESSION,155,1,WRITE,INSERT,TABLE,public.ri_fk,"INSERT INTO ri_fk VALUES (1), (2)",<none>,2 +BEGIN; +SET CONSTRAINTS ALL DEFERRED; +INSERT INTO ri_fk VALUES (1), (2); +NOTICE: AUDIT: SESSION,156,1,WRITE,INSERT,TABLE,public.ri_fk,"INSERT INTO ri_fk VALUES (1), (2)",<none>,2 +SET CONSTRAINTS ALL IMMEDIATE; +COMMIT; +COPY ri_fk FROM stdin; +NOTICE: AUDIT: SESSION,157,1,WRITE,COPY,TABLE,public.ri_fk,COPY ri_fk FROM stdin,<none>,2 +CREATE PROCEDURE ri_proc() LANGUAGE plpgsql AS $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; + INSERT INTO ri_fk VALUES (2); +END $$; +NOTICE: AUDIT: SESSION,158,1,DDL,CREATE PROCEDURE,,,"CREATE PROCEDURE ri_proc() LANGUAGE plpgsql AS $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; + INSERT INTO ri_fk VALUES (2); +END $$",<none>,0 +CALL ri_proc(); +NOTICE: AUDIT: SESSION,159,1,FUNCTION,EXECUTE,FUNCTION,public.ri_proc,CALL ri_proc(),<none>,0 +NOTICE: AUDIT: SESSION,159,2,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (1),<none>,1 +NOTICE: AUDIT: SESSION,159,3,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (2),<none>,1 +DO $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; +END $$; +NOTICE: AUDIT: SESSION,160,1,FUNCTION,DO,,,"DO $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; +END $$",<none>,0 +NOTICE: AUDIT: SESSION,160,2,WRITE,INSERT,TABLE,public.ri_fk,INSERT INTO ri_fk VALUES (1),<none>,1 +DROP PROCEDURE ri_proc(); +NOTICE: AUDIT: SESSION,161,1,DDL,DROP PROCEDURE,,,DROP PROCEDURE ri_proc(),<none>,0 +DROP TABLE ri_fk; +NOTICE: AUDIT: SESSION,162,1,DDL,DROP TABLE,,,DROP TABLE ri_fk,<none>,0 +DROP TABLE ri_pk; +NOTICE: AUDIT: SESSION,163,1,DDL,DROP TABLE,,,DROP TABLE ri_pk,<none>,0 +-- -- Change configuration of user 1 so that full statements are not logged \connect - :current_user ALTER ROLE regress_user1 RESET pgaudit.log_relation; diff --git a/pgaudit.c b/pgaudit.c index 60e508c..59a9c75 100644 --- a/pgaudit.c +++ b/pgaudit.c @@ -264,6 +264,7 @@ typedef struct List *rangeTabls; /* Tables in query tracking rows */ List *permInfos; /* Permission info rows for each table involved in the query */ + bool permsPending; /* Next permission check is this event's own */ } AuditEvent; /* @@ -1485,6 +1486,7 @@ pgaudit_ExecutorStart_hook(QueryDesc *queryDesc, int eflags) { /* Push the audit event onto the stack */ stackItem = stack_push(); + stackItem->auditEvent.permsPending = true; /* Initialize command using queryDesc->operation */ switch (queryDesc->operation) @@ -1549,6 +1551,9 @@ pgaudit_ExecutorStart_hook(QueryDesc *queryDesc, int eflags) */ if (stackItem) { + /* Checks made after ExecutorStart are not this event's own */ + stackItem->auditEvent.permsPending = false; + MemoryContextSetParent(stackItem->contextAudit, queryDesc->estate->es_query_cxt); @@ -1566,6 +1571,25 @@ pgaudit_ExecutorCheckPerms_hook(List *rangeTabls, List *permInfos, bool ereport_on_violation) { + /* + * Only audit the top stack item's own check, i.e. the first one made by its + * ExecutorStart or by COPY. Consume it even when not auditing so a later + * check cannot be mistaken for it. + * + * Any other check runs on behalf of something other than the statement on + * top of the stack, so auditing it would attribute the relations to that + * statement, or fail when the stack is empty. Since PostgreSQL 19 the RI + * fast path checks from FK triggers, which may fire after the statement's + * own check or be deferred to COMMIT. RI_Initial_Check() also checks while + * the utility command's stack item is on top. The queries these go on to + * run, if any, are audited normally. + */ + bool ownCheck = auditEventStack != NULL && + auditEventStack->auditEvent.permsPending; + + if (ownCheck) + auditEventStack->auditEvent.permsPending = false; + /* * Only do audit work when an executor-driven class is enabled or object * auditing via pgaudit.role might apply. This short-circuits the role @@ -1575,13 +1599,9 @@ pgaudit_ExecutorCheckPerms_hook(List *rangeTabls, * Also skip checks that are not enforcing. A caller passing * ereport_on_violation false is asking whether it may proceed, not * executing a statement against the relations, so there is nothing to - * audit. RI_Initial_Check() does this to decide whether it can validate a - * foreign key with a single query, and it runs while the utility command's - * own stack item is on top of the stack, so auditing the check would - * attribute the referenced relations to that command. The validation query - * it goes on to run is audited normally. + * audit. */ - if (audit_executor_enabled() && ereport_on_violation && + if (ownCheck && audit_executor_enabled() && ereport_on_violation && !IsAbortedTransactionBlockState() && !IsParallelWorker()) { /* Get the audit oid if the role exists */ @@ -1591,7 +1611,7 @@ pgaudit_ExecutorCheckPerms_hook(List *rangeTabls, if (auditOid != InvalidOid || auditLogBitmap != 0) { /* If auditLogRows is on, wait for rows processed to be set */ - if (auditLogRows && auditEventStack != NULL) + if (auditLogRows) { /* Check if the top item is SELECT/INSERT for CREATE TABLE AS */ if (auditEventStack->auditEvent.commandTag == T_SelectStmt && @@ -1618,10 +1638,7 @@ pgaudit_ExecutorCheckPerms_hook(List *rangeTabls, } } else - { - STACK_NOT_EMPTY(); log_select_dml(auditOid, rangeTabls, permInfos); - } } } @@ -1745,6 +1762,8 @@ pgaudit_ProcessUtility_hook(PlannedStmt *pstmt, stackItem->auditEvent.logStmtLevel = GetCommandLogLevel(pstmt->utilityStmt); stackItem->auditEvent.commandTag = nodeTag(pstmt->utilityStmt); stackItem->auditEvent.command = CreateCommandTag(pstmt->utilityStmt); + stackItem->auditEvent.permsPending = + stackItem->auditEvent.commandTag == T_CopyStmt; command_text_set(&stackItem->auditEvent, queryString, pstmt->stmt_location, pstmt->stmt_len); diff --git a/sql/pgaudit.sql b/sql/pgaudit.sql index cd496cf..3b9f6e4 100644 --- a/sql/pgaudit.sql +++ b/sql/pgaudit.sql @@ -994,6 +994,50 @@ ALTER TABLE fk_fk VALIDATE CONSTRAINT fk_fk_id_fkey_nv; DROP TABLE fk_fk; DROP TABLE fk_pk; +-- +-- Test that FK checks made by the RI fast path are not logged against the +-- statement. Since PostgreSQL 19 the fast path checks permissions from the FK +-- trigger, which runs after the statement's own check or, when the constraint +-- is deferred, at COMMIT. A top-level COMMIT runs with the stack empty, while +-- one made by a procedure or DO block runs with the CALL or DO on top. +CREATE TABLE ri_pk (id int PRIMARY KEY); +CREATE TABLE ri_fk (id int REFERENCES ri_pk (id) DEFERRABLE); +INSERT INTO ri_pk VALUES (1), (2); + +INSERT INTO ri_fk VALUES (1); + +BEGIN; +SET CONSTRAINTS ALL DEFERRED; +INSERT INTO ri_fk VALUES (2); +COMMIT; + +COPY ri_fk FROM stdin; +1 +2 +\. + +CREATE PROCEDURE ri_proc() LANGUAGE plpgsql AS $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; + INSERT INTO ri_fk VALUES (2); +END $$; + +CALL ri_proc(); + +DO $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; +END $$; + +DROP PROCEDURE ri_proc(); + +DROP TABLE ri_fk; +DROP TABLE ri_pk; + -- -- Test rows retrived or affected by statements \connect - :current_user @@ -1659,6 +1703,51 @@ CREATE INDEX h_idx ON h (x); DROP INDEX h_idx; DROP TABLE h; +-- +-- Test that FK checks made by the RI fast path are not logged against the +-- statement. Since PostgreSQL 19 the fast path checks permissions from the FK +-- trigger, which runs after the statement's own check or, when the constraint +-- is deferred, at SET CONSTRAINTS ... IMMEDIATE or COMMIT, including a COMMIT +-- made by a procedure or DO block. +CREATE TABLE ri_pk (id int PRIMARY KEY); +CREATE TABLE ri_fk (id int REFERENCES ri_pk (id) DEFERRABLE); +INSERT INTO ri_pk VALUES (1), (2); + +INSERT INTO ri_fk VALUES (1), (2); + +BEGIN; +SET CONSTRAINTS ALL DEFERRED; +INSERT INTO ri_fk VALUES (1), (2); +SET CONSTRAINTS ALL IMMEDIATE; +COMMIT; + +COPY ri_fk FROM stdin; +1 +2 +\. + +CREATE PROCEDURE ri_proc() LANGUAGE plpgsql AS $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; + INSERT INTO ri_fk VALUES (2); +END $$; + +CALL ri_proc(); + +DO $$ +BEGIN + SET CONSTRAINTS ALL DEFERRED; + INSERT INTO ri_fk VALUES (1); + COMMIT; +END $$; + +DROP PROCEDURE ri_proc(); + +DROP TABLE ri_fk; +DROP TABLE ri_pk; + -- -- Change configuration of user 1 so that full statements are not logged \connect - :current_user
^ permalink raw [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-01 21:23  "adeshthack (@adeshthack)" <noreply+adeshthack@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: adeshthack (@adeshthack) @ 2026-10-01 21:23 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

is there any update on this?

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-05 16:14  "zjyzh (@zjyzh)" <noreply+zjyzh@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: zjyzh (@zjyzh) @ 2026-10-05 16:14 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

Hi @dwsteele can you review/approve this PR? thanks

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-06 06:36  "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: dwsteele (@dwsteele) @ 2026-10-06 06:36 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

@dwsteele assigned @dwsteele.

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-06 06:36  "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: dwsteele (@dwsteele) @ 2026-10-06 06:36 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

@dwsteele applied label "bug".

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-06 06:36  "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: dwsteele (@dwsteele) @ 2026-10-06 06:36 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

@dwsteele requested a review.

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334 (approved): Only audit a statement's own permission check.
@ 2026-10-06 07:13  "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: dwsteele (@dwsteele) @ 2026-10-06 07:13 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

This looks good to me. I made some fixes/improvements in a1526f6e3aca, 820c3426a, and dbad44fc. If those look good then this should be ready to commit.

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-21 13:27  "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: dwsteele (@dwsteele) @ 2026-10-21 13:27 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

@zjyzh I'm waiting on your approval of my changes before merging. See my last comment.

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-27 09:09  "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: dwsteele (@dwsteele) @ 2026-10-27 09:09 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

@dwsteele merged this (commit 4313556d245d).

^ permalink  raw  [nested|flat] 10+ messages in thread

* Re: [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check.
@ 2026-10-27 09:09  "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
  8 siblings, 0 replies; 10+ messages in thread

From: dwsteele (@dwsteele) @ 2026-10-27 09:09 UTC (permalink / raw)
  To: pgaudit/pgaudit <noreply+pgaudit-pgaudit@github.com>

@dwsteele closed this.

^ permalink  raw  [nested|flat] 10+ messages in thread


end of thread, other threads:[~2026-10-27 09:09 UTC | newest]

Thread overview: 10+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2026-09-30 18:10 [pgaudit/pgaudit] PR #334: Only audit a statement's own permission check. "zjyzh (@zjyzh)" <noreply+zjyzh@github.com>
2026-10-01 21:23 ` "adeshthack (@adeshthack)" <noreply+adeshthack@github.com>
2026-10-05 16:14 ` "zjyzh (@zjyzh)" <noreply+zjyzh@github.com>
2026-10-06 06:36 ` "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
2026-10-06 06:36 ` "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
2026-10-06 06:36 ` "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
2026-10-06 07:13 ` Re: [pgaudit/pgaudit] PR #334 (approved): Only audit a statement's own permission check. "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
2026-10-21 13:27 ` "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
2026-10-27 09:09 ` "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>
2026-10-27 09:09 ` "dwsteele (@dwsteele)" <noreply+dwsteele@github.com>

This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox