agora inbox for pgsql-hackers@postgresql.org
help / color / mirror / Atom feedRe: explain analyze rows=%.0f
464+ messages / 7 participants
[nested] [flat]
* Re: explain analyze rows=%.0f
@ 2009-06-02 03:30 Ron Mayer <rm_pg@cheapcomplexdevices.com>
0 siblings, 1 reply; 464+ messages in thread
From: Ron Mayer @ 2009-06-02 03:30 UTC (permalink / raw)
To: Euler Taveira de Oliveira <euler@timbira.com>; +Cc: Robert Haas <robertmhaas@gmail.com>; pgsql-hackers
Euler Taveira de Oliveira wrote:
> Robert Haas escreveu:
>> ...EXPLAIN ANALYZE reports the number of rows as an integer... Any
>> chance we could reconsider this decision? I often find myself wanting
>> to know the value that is here called ntuples, but rounding
>> ntuples/nloops off to the nearest integer loses too much precision.
>>
> Don't you think is too strange having, for example, 6.67 rows? I would confuse
> users and programs that parses the EXPLAIN output. However, I wouldn't object
I don't think it's that confusing. If it says "0.1 rows", I imagine most
people would infer that this means "typically 0, but sometimes 1 or a few" rows.
What I'd find strange about "6.67 rows" in your example is more that on
the estimated rows side, it seems to imply an unrealistically precise estimate
in the same way that "667 rows" would seem unrealistically precise to me.
Maybe rounding to 2 significant digits would reduce confusion?
^ permalink raw reply [nested|flat] 464+ messages in thread
* Re: explain analyze rows=%.0f
@ 2009-06-02 13:41 Simon Riggs <simon@2ndQuadrant.com>
parent: Ron Mayer <rm_pg@cheapcomplexdevices.com>
0 siblings, 1 reply; 464+ messages in thread
From: Simon Riggs @ 2009-06-02 13:41 UTC (permalink / raw)
To: Ron Mayer <rm_pg@cheapcomplexdevices.com>; +Cc: Euler Taveira de Oliveira <euler@timbira.com>; Robert Haas <robertmhaas@gmail.com>; pgsql-hackers
On Mon, 2009-06-01 at 20:30 -0700, Ron Mayer wrote:
> What I'd find strange about "6.67 rows" in your example is more that on
> the estimated rows side, it seems to imply an unrealistically precise estimate
> in the same way that "667 rows" would seem unrealistically precise to me.
> Maybe rounding to 2 significant digits would reduce confusion?
You're right that the number of significant digits already exceeds the
true accuracy of the computation. I think what Robert wants to see is
the exact value used in the calc, so the estimates can be checked more
thoroughly than is currently possible.
--
Simon Riggs www.2ndQuadrant.com
PostgreSQL Training, Services and Support
^ permalink raw reply [nested|flat] 464+ messages in thread
* Re: explain analyze rows=%.0f
@ 2009-06-02 14:06 Robert Haas <robertmhaas@gmail.com>
parent: Simon Riggs <simon@2ndQuadrant.com>
0 siblings, 1 reply; 464+ messages in thread
From: Robert Haas @ 2009-06-02 14:06 UTC (permalink / raw)
To: Simon Riggs <simon@2ndQuadrant.com>; +Cc: Ron Mayer <rm_pg@cheapcomplexdevices.com>; Euler Taveira de Oliveira <euler@timbira.com>; pgsql-hackers
On Jun 2, 2009, at 9:41 AM, Simon Riggs <simon@2ndQuadrant.com> wrote:
>
> On Mon, 2009-06-01 at 20:30 -0700, Ron Mayer wrote:
>
>> What I'd find strange about "6.67 rows" in your example is more
>> that on
>> the estimated rows side, it seems to imply an unrealistically
>> precise estimate
>> in the same way that "667 rows" would seem unrealistically precise
>> to me.
>> Maybe rounding to 2 significant digits would reduce confusion?
>
> You're right that the number of significant digits already exceeds the
> true accuracy of the computation. I think what Robert wants to see is
> the exact value used in the calc, so the estimates can be checked more
> thoroughly than is currently possible.
Bingo.
...Robert
^ permalink raw reply [nested|flat] 464+ messages in thread
* Re: explain analyze rows=%.0f
@ 2009-06-02 14:38 Tom Lane <tgl@sss.pgh.pa.us>
parent: Robert Haas <robertmhaas@gmail.com>
0 siblings, 1 reply; 464+ messages in thread
From: Tom Lane @ 2009-06-02 14:38 UTC (permalink / raw)
To: Robert Haas <robertmhaas@gmail.com>; +Cc: Simon Riggs <simon@2ndQuadrant.com>; Ron Mayer <rm_pg@cheapcomplexdevices.com>; Euler Taveira de Oliveira <euler@timbira.com>; pgsql-hackers
Robert Haas <robertmhaas@gmail.com> writes:
> On Jun 2, 2009, at 9:41 AM, Simon Riggs <simon@2ndQuadrant.com> wrote:
>> You're right that the number of significant digits already exceeds the
>> true accuracy of the computation. I think what Robert wants to see is
>> the exact value used in the calc, so the estimates can be checked more
>> thoroughly than is currently possible.
> Bingo.
Uh, the planner's estimate *is* an integer. What was under discussion
(I thought) was showing some fractional digits in the case where EXPLAIN
ANALYZE is outputting a measured row count that is an average over
multiple loops, and therefore isn't necessarily an integer. In that
case the measured value can be considered arbitrarily precise --- though
I think in practice one or two fractional digits would be plenty.
regards, tom lane
^ permalink raw reply [nested|flat] 464+ messages in thread
* Re: explain analyze rows=%.0f
@ 2009-06-02 14:56 Robert Haas <robertmhaas@gmail.com>
parent: Tom Lane <tgl@sss.pgh.pa.us>
0 siblings, 0 replies; 464+ messages in thread
From: Robert Haas @ 2009-06-02 14:56 UTC (permalink / raw)
To: Tom Lane <tgl@sss.pgh.pa.us>; +Cc: Simon Riggs <simon@2ndQuadrant.com>; Ron Mayer <rm_pg@cheapcomplexdevices.com>; Euler Taveira de Oliveira <euler@timbira.com>; pgsql-hackers
...Robert
On Jun 2, 2009, at 10:38 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> On Jun 2, 2009, at 9:41 AM, Simon Riggs <simon@2ndQuadrant.com>
>> wrote:
>>> You're right that the number of significant digits already exceeds
>>> the
>>> true accuracy of the computation. I think what Robert wants to see
>>> is
>>> the exact value used in the calc, so the estimates can be checked
>>> more
>>> thoroughly than is currently possible.
>
>> Bingo.
>
> Uh, the planner's estimate *is* an integer. What was under discussion
> (I thought) was showing some fractional digits in the case where
> EXPLAIN
> ANALYZE is outputting a measured row count that is an average over
> multiple loops, and therefore isn't necessarily an integer. In that
> case the measured value can be considered arbitrarily precise ---
> though
> I think in practice one or two fractional digits would be plenty.
We're in violent agreement here.
...Robert
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v1 1/2] Report wait event for cost-based vacuum delay
@ 2020-03-21 01:47 Justin Pryzby <pryzbyj@telsasoft.com>
0 siblings, 0 replies; 464+ messages in thread
From: Justin Pryzby @ 2020-03-21 01:47 UTC (permalink / raw)
---
doc/src/sgml/monitoring.sgml | 2 ++
src/backend/commands/vacuum.c | 2 ++
src/backend/postmaster/pgstat.c | 3 +++
src/include/pgstat.h | 3 ++-
4 files changed, 9 insertions(+), 1 deletion(-)
diff --git a/doc/src/sgml/monitoring.sgml b/doc/src/sgml/monitoring.sgml
index 5bffdcce10..46c99a55b7 100644
--- a/doc/src/sgml/monitoring.sgml
+++ b/doc/src/sgml/monitoring.sgml
@@ -1507,6 +1507,8 @@ postgres 27093 0.0 0.0 30096 2752 ? Ss 11:34 0:00 postgres: ser
(<filename>pg_wal</filename>, archive or stream) before trying
again to retrieve WAL data, at recovery.
</entry>
+ <entry><literal>VacuumDelay</literal></entry>
+ <entry>Waiting in a cost-based vacuum delay point.</entry>
</row>
<row>
<entry morerows="68"><literal>IO</literal></entry>
diff --git a/src/backend/commands/vacuum.c b/src/backend/commands/vacuum.c
index d625d17bf4..59731d687f 100644
--- a/src/backend/commands/vacuum.c
+++ b/src/backend/commands/vacuum.c
@@ -2019,7 +2019,9 @@ vacuum_delay_point(void)
if (msec > VacuumCostDelay * 4)
msec = VacuumCostDelay * 4;
+ pgstat_report_wait_start(WAIT_EVENT_VACUUM_DELAY);
pg_usleep((long) (msec * 1000));
+ pgstat_report_wait_end();
VacuumCostBalance = 0;
diff --git a/src/backend/postmaster/pgstat.c b/src/backend/postmaster/pgstat.c
index d29c211a76..742ec07b59 100644
--- a/src/backend/postmaster/pgstat.c
+++ b/src/backend/postmaster/pgstat.c
@@ -3824,6 +3824,9 @@ pgstat_get_wait_timeout(WaitEventTimeout w)
case WAIT_EVENT_RECOVERY_RETRIEVE_RETRY_INTERVAL:
event_name = "RecoveryRetrieveRetryInterval";
break;
+ case WAIT_EVENT_VACUUM_DELAY:
+ event_name = "VacuumDelay";
+ break;
/* no default case, so that compiler will warn */
}
diff --git a/src/include/pgstat.h b/src/include/pgstat.h
index 851d0a7246..4db40e23cc 100644
--- a/src/include/pgstat.h
+++ b/src/include/pgstat.h
@@ -848,7 +848,8 @@ typedef enum
WAIT_EVENT_BASE_BACKUP_THROTTLE = PG_WAIT_TIMEOUT,
WAIT_EVENT_PG_SLEEP,
WAIT_EVENT_RECOVERY_APPLY_DELAY,
- WAIT_EVENT_RECOVERY_RETRIEVE_RETRY_INTERVAL
+ WAIT_EVENT_RECOVERY_RETRIEVE_RETRY_INTERVAL,
+ WAIT_EVENT_VACUUM_DELAY,
} WaitEventTimeout;
/* ----------
--
2.17.0
--9jxsPFA5p3P2qPhR
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="v1-0002-vacuum-to-report-time-spent-in-cost-based-delay.patch"
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v1 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use CREATE OR REPLACE FUNCTION to handle the
optional pretty argument.
---
src/backend/catalog/system_functions.sql | 7 +++++++
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 5 +----
3 files changed, 8 insertions(+), 21 deletions(-)
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index 1f89dcc7908..81d210a9c45 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -692,6 +692,13 @@ LANGUAGE INTERNAL
PARALLEL SAFE
AS 'pg_get_constraintdef';
+CREATE OR REPLACE FUNCTION
+ pg_get_expr(expr pg_node_tree, relation oid, pretty bool DEFAULT false)
+RETURNS TEXT
+LANGUAGE INTERNAL
+PARALLEL SAFE
+AS 'pg_get_expr';
+
--
-- The default permissions for functions mean that anyone can execute them.
-- A number of functions shouldn't be executable by just anyone, but rather
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 416144c49f5..9effa02fa2e 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2573,23 +2573,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index a9431ea4037..10c5286bd7d 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3977,9 +3977,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8502,7 +8499,7 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.51.2
--IRGsBYp1UN8d6WrT
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v1-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v2 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use CREATE OR REPLACE FUNCTION to handle the
optional pretty argument.
---
src/backend/catalog/system_functions.sql | 7 +++++++
src/backend/commands/tablecmds.c | 4 ++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 5 +----
4 files changed, 10 insertions(+), 23 deletions(-)
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index 1824faab231..a6f7cdf3a36 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -692,6 +692,13 @@ LANGUAGE INTERNAL
PARALLEL SAFE
AS 'pg_get_constraintdef';
+CREATE OR REPLACE FUNCTION
+ pg_get_expr(expr pg_node_tree, relation oid, pretty bool DEFAULT false)
+RETURNS TEXT
+LANGUAGE INTERNAL
+PARALLEL SAFE
+AS 'pg_get_expr';
+
--
-- The default permissions for functions mean that anyone can execute them.
-- A number of functions shouldn't be executable by just anyone, but rather
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index f976c0e5c7e..12c4ce08437 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17480,8 +17480,8 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid), false);
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 9faf340f0c6..305e3e80c41 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2573,23 +2573,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 96eb7baf8e0..3a0be2d4b2a 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3986,9 +3986,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8518,7 +8515,7 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.43.0
--3/nz5wg6+DYwHsLU
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="v2-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch"
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v3 6/7] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use CREATE OR REPLACE FUNCTION to handle the
optional pretty argument.
---
src/backend/catalog/system_functions.sql | 7 +++++++
src/backend/commands/tablecmds.c | 4 ++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 5 +----
src/include/catalog/pg_retired.dat | 3 ++-
5 files changed, 12 insertions(+), 24 deletions(-)
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index 1824faab231..a6f7cdf3a36 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -692,6 +692,13 @@ LANGUAGE INTERNAL
PARALLEL SAFE
AS 'pg_get_constraintdef';
+CREATE OR REPLACE FUNCTION
+ pg_get_expr(expr pg_node_tree, relation oid, pretty bool DEFAULT false)
+RETURNS TEXT
+LANGUAGE INTERNAL
+PARALLEL SAFE
+AS 'pg_get_expr';
+
--
-- The default permissions for functions mean that anyone can execute them.
-- A number of functions shouldn't be executable by just anyone, but rather
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index f976c0e5c7e..12c4ce08437 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17480,8 +17480,8 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid), false);
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 4f2c93f1ee2..d87b361a093 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2573,23 +2573,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index ffec11c5539..4304ab220ba 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3986,9 +3986,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8518,7 +8515,7 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
diff --git a/src/include/catalog/pg_retired.dat b/src/include/catalog/pg_retired.dat
index 90928a9cb00..8b5f58850f6 100644
--- a/src/include/catalog/pg_retired.dat
+++ b/src/include/catalog/pg_retired.dat
@@ -21,6 +21,7 @@
{ oid => '1573', proname => 'pg_get_ruledef' },
{ oid => '1640', proname => 'pg_get_viewdef' },
{ oid => '1641', proname => 'pg_get_viewdef' },
-{ oid => '1643', proname => 'pg_get_indexdef' }
+{ oid => '1643', proname => 'pg_get_indexdef' },
+{ oid => '1716', proname => 'pg_get_expr' }
]
--
2.43.0
--zd8Z8rlPOcTi77n/
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="v3-0007-Handle-pg_get_triggerdef-default-args-in-system_f.patch"
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v7 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 9 ++++-----
3 files changed, 7 insertions(+), 24 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index eec09ba1ded..73eb6322daf 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17695,8 +17695,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 2d39bce7fd7..707f83d4310 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 39139fd9e3b..855529509de 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4006,9 +4006,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8638,9 +8635,11 @@
pronargdefaults => '1', proargdefaults => '{NULL}',
prosrc => 'pg_get_database_ddl' },
{ oid => '2509',
- descr => 'deparse an encoded expression with pretty-print option',
+ descr => 'deparse an encoded expression',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.52.0
--GzYugJnip6WHHvTk
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v7-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v8 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 9 ++++-----
3 files changed, 7 insertions(+), 24 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index a1845240a98..f063f744f6d 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17730,8 +17730,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index d83be8cf77d..e60478677e0 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 7e452aa8c27..71bd2aaf3d9 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4006,9 +4006,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8605,9 +8602,11 @@
proargmodes => '{i,v}', proargdefaults => '{NULL}',
prosrc => 'pg_get_database_ddl' },
{ oid => '2509',
- descr => 'deparse an encoded expression with pretty-print option',
+ descr => 'deparse an encoded expression',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.53.0
--iWwo79nqhOvOs5kf
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v8-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v8.1 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 9 ++++-----
3 files changed, 7 insertions(+), 24 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index a1845240a98..f063f744f6d 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17730,8 +17730,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index d83be8cf77d..e60478677e0 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index e610a6213ec..df43469be3d 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4006,9 +4006,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8605,9 +8602,11 @@
proargmodes => '{i,v}', proargdefaults => '{NULL}',
prosrc => 'pg_get_database_ddl' },
{ oid => '2509',
- descr => 'deparse an encoded expression with pretty-print option',
+ descr => 'deparse an encoded expression',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.53.0
--AcCatzXgbvyipyEy
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v8.1-0006-Handle-pg_get_triggerdef-default-args-in-system.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v3 6/7] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use CREATE OR REPLACE FUNCTION to handle the
optional pretty argument.
---
src/backend/catalog/system_functions.sql | 7 +++++++
src/backend/commands/tablecmds.c | 4 ++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 5 +----
src/include/catalog/pg_retired.dat | 3 ++-
5 files changed, 12 insertions(+), 24 deletions(-)
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index 1824faab231..a6f7cdf3a36 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -692,6 +692,13 @@ LANGUAGE INTERNAL
PARALLEL SAFE
AS 'pg_get_constraintdef';
+CREATE OR REPLACE FUNCTION
+ pg_get_expr(expr pg_node_tree, relation oid, pretty bool DEFAULT false)
+RETURNS TEXT
+LANGUAGE INTERNAL
+PARALLEL SAFE
+AS 'pg_get_expr';
+
--
-- The default permissions for functions mean that anyone can execute them.
-- A number of functions shouldn't be executable by just anyone, but rather
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index f976c0e5c7e..12c4ce08437 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17480,8 +17480,8 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid), false);
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 4f2c93f1ee2..d87b361a093 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2573,23 +2573,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index ffec11c5539..4304ab220ba 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3986,9 +3986,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8518,7 +8515,7 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
diff --git a/src/include/catalog/pg_retired.dat b/src/include/catalog/pg_retired.dat
index 90928a9cb00..8b5f58850f6 100644
--- a/src/include/catalog/pg_retired.dat
+++ b/src/include/catalog/pg_retired.dat
@@ -21,6 +21,7 @@
{ oid => '1573', proname => 'pg_get_ruledef' },
{ oid => '1640', proname => 'pg_get_viewdef' },
{ oid => '1641', proname => 'pg_get_viewdef' },
-{ oid => '1643', proname => 'pg_get_indexdef' }
+{ oid => '1643', proname => 'pg_get_indexdef' },
+{ oid => '1716', proname => 'pg_get_expr' }
]
--
2.43.0
--zd8Z8rlPOcTi77n/
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="v3-0007-Handle-pg_get_triggerdef-default-args-in-system_f.patch"
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v4 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 7 +++----
3 files changed, 6 insertions(+), 23 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index b04b0dbd2a0..a523a512949 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17477,8 +17477,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 03550f0d80b..a70807e84ca 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2574,23 +2574,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 1d1b24a8f4c..69c1e5bb264 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3987,9 +3987,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8531,7 +8528,9 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.43.0
--eEEy3EHc57lA8spa
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="v4-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch"
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v5 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 7 +++----
3 files changed, 6 insertions(+), 23 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index 0ce2e81f9c2..fad4202b1ff 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17694,8 +17694,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 1afa2531f30..972f89eafbc 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index e148b1ad8e3..f36f4625e50 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4000,9 +4000,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8595,7 +8592,9 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.52.0
--m23X5uHFNxthGTU3
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v5-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v8 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 9 ++++-----
3 files changed, 7 insertions(+), 24 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index a1845240a98..f063f744f6d 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17730,8 +17730,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index d83be8cf77d..e60478677e0 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 7e452aa8c27..71bd2aaf3d9 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4006,9 +4006,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8605,9 +8602,11 @@
proargmodes => '{i,v}', proargdefaults => '{NULL}',
prosrc => 'pg_get_database_ddl' },
{ oid => '2509',
- descr => 'deparse an encoded expression with pretty-print option',
+ descr => 'deparse an encoded expression',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.53.0
--iWwo79nqhOvOs5kf
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v8-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v7 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 9 ++++-----
3 files changed, 7 insertions(+), 24 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index eec09ba1ded..73eb6322daf 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17695,8 +17695,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 2d39bce7fd7..707f83d4310 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 39139fd9e3b..855529509de 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4006,9 +4006,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8638,9 +8635,11 @@
pronargdefaults => '1', proargdefaults => '{NULL}',
prosrc => 'pg_get_database_ddl' },
{ oid => '2509',
- descr => 'deparse an encoded expression with pretty-print option',
+ descr => 'deparse an encoded expression',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.52.0
--GzYugJnip6WHHvTk
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v7-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v5 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 7 +++----
3 files changed, 6 insertions(+), 23 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index 0ce2e81f9c2..fad4202b1ff 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17694,8 +17694,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 1afa2531f30..972f89eafbc 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index e148b1ad8e3..f36f4625e50 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4000,9 +4000,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8595,7 +8592,9 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.52.0
--m23X5uHFNxthGTU3
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v5-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v1 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use CREATE OR REPLACE FUNCTION to handle the
optional pretty argument.
---
src/backend/catalog/system_functions.sql | 7 +++++++
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 5 +----
3 files changed, 8 insertions(+), 21 deletions(-)
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index 1f89dcc7908..81d210a9c45 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -692,6 +692,13 @@ LANGUAGE INTERNAL
PARALLEL SAFE
AS 'pg_get_constraintdef';
+CREATE OR REPLACE FUNCTION
+ pg_get_expr(expr pg_node_tree, relation oid, pretty bool DEFAULT false)
+RETURNS TEXT
+LANGUAGE INTERNAL
+PARALLEL SAFE
+AS 'pg_get_expr';
+
--
-- The default permissions for functions mean that anyone can execute them.
-- A number of functions shouldn't be executable by just anyone, but rather
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 416144c49f5..9effa02fa2e 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2573,23 +2573,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index a9431ea4037..10c5286bd7d 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3977,9 +3977,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8502,7 +8499,7 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.51.2
--IRGsBYp1UN8d6WrT
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v1-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v2 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use CREATE OR REPLACE FUNCTION to handle the
optional pretty argument.
---
src/backend/catalog/system_functions.sql | 7 +++++++
src/backend/commands/tablecmds.c | 4 ++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 5 +----
4 files changed, 10 insertions(+), 23 deletions(-)
diff --git a/src/backend/catalog/system_functions.sql b/src/backend/catalog/system_functions.sql
index 1824faab231..a6f7cdf3a36 100644
--- a/src/backend/catalog/system_functions.sql
+++ b/src/backend/catalog/system_functions.sql
@@ -692,6 +692,13 @@ LANGUAGE INTERNAL
PARALLEL SAFE
AS 'pg_get_constraintdef';
+CREATE OR REPLACE FUNCTION
+ pg_get_expr(expr pg_node_tree, relation oid, pretty bool DEFAULT false)
+RETURNS TEXT
+LANGUAGE INTERNAL
+PARALLEL SAFE
+AS 'pg_get_expr';
+
--
-- The default permissions for functions mean that anyone can execute them.
-- A number of functions shouldn't be executable by just anyone, but rather
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index f976c0e5c7e..12c4ce08437 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17480,8 +17480,8 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid), false);
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 9faf340f0c6..305e3e80c41 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2573,23 +2573,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 96eb7baf8e0..3a0be2d4b2a 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3986,9 +3986,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8518,7 +8515,7 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.43.0
--3/nz5wg6+DYwHsLU
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="v2-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch"
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v8.1 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 9 ++++-----
3 files changed, 7 insertions(+), 24 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index a1845240a98..f063f744f6d 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17730,8 +17730,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index d83be8cf77d..e60478677e0 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2932,23 +2932,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index e610a6213ec..df43469be3d 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -4006,9 +4006,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8605,9 +8602,11 @@
proargmodes => '{i,v}', proargdefaults => '{NULL}',
prosrc => 'pg_get_database_ddl' },
{ oid => '2509',
- descr => 'deparse an encoded expression with pretty-print option',
+ descr => 'deparse an encoded expression',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.53.0
--AcCatzXgbvyipyEy
Content-Type: text/plain; charset=us-ascii
Content-Disposition: attachment;
filename=v8.1-0006-Handle-pg_get_triggerdef-default-args-in-system.patch
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH v4 5/6] Handle pg_get_expr default args in system_functions.sql
@ 2025-12-09 19:17 Mark Wong <markwkm@gmail.com>
0 siblings, 0 replies; 464+ messages in thread
From: Mark Wong @ 2025-12-09 19:17 UTC (permalink / raw)
Modernize pg_get_expr to use proargdefaults to handle the optional
pretty argument. That also means any direct function calls now needs to
set the pretty parameter.
---
src/backend/commands/tablecmds.c | 5 +++--
src/backend/utils/adt/ruleutils.c | 17 -----------------
src/include/catalog/pg_proc.dat | 7 +++----
3 files changed, 6 insertions(+), 23 deletions(-)
diff --git a/src/backend/commands/tablecmds.c b/src/backend/commands/tablecmds.c
index b04b0dbd2a0..a523a512949 100644
--- a/src/backend/commands/tablecmds.c
+++ b/src/backend/commands/tablecmds.c
@@ -17477,8 +17477,9 @@ decompile_conbin(HeapTuple contup, TupleDesc tupdesc)
if (isnull)
elog(ERROR, "null conbin for constraint %u", con->oid);
- expr = DirectFunctionCall2(pg_get_expr, attr,
- ObjectIdGetDatum(con->conrelid));
+ expr = DirectFunctionCall3(pg_get_expr, attr,
+ ObjectIdGetDatum(con->conrelid),
+ BoolGetDatum(false));
return TextDatumGetCString(expr);
}
diff --git a/src/backend/utils/adt/ruleutils.c b/src/backend/utils/adt/ruleutils.c
index 03550f0d80b..a70807e84ca 100644
--- a/src/backend/utils/adt/ruleutils.c
+++ b/src/backend/utils/adt/ruleutils.c
@@ -2574,23 +2574,6 @@ decompile_column_index_array(Datum column_index_array, Oid relId,
*/
Datum
pg_get_expr(PG_FUNCTION_ARGS)
-{
- text *expr = PG_GETARG_TEXT_PP(0);
- Oid relid = PG_GETARG_OID(1);
- text *result;
- int prettyFlags;
-
- prettyFlags = PRETTYFLAG_INDENT;
-
- result = pg_get_expr_worker(expr, relid, prettyFlags);
- if (result)
- PG_RETURN_TEXT_P(result);
- else
- PG_RETURN_NULL();
-}
-
-Datum
-pg_get_expr_ext(PG_FUNCTION_ARGS)
{
text *expr = PG_GETARG_TEXT_PP(0);
Oid relid = PG_GETARG_OID(1);
diff --git a/src/include/catalog/pg_proc.dat b/src/include/catalog/pg_proc.dat
index 1d1b24a8f4c..69c1e5bb264 100644
--- a/src/include/catalog/pg_proc.dat
+++ b/src/include/catalog/pg_proc.dat
@@ -3987,9 +3987,6 @@
{ oid => '1662', descr => 'trigger description',
proname => 'pg_get_triggerdef', provolatile => 's', prorettype => 'text',
proargtypes => 'oid', prosrc => 'pg_get_triggerdef' },
-{ oid => '1716', descr => 'deparse an encoded expression',
- proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid', prosrc => 'pg_get_expr' },
{ oid => '1665', descr => 'name of sequence for a serial column',
proname => 'pg_get_serial_sequence', provolatile => 's', prorettype => 'text',
proargtypes => 'text text', prosrc => 'pg_get_serial_sequence' },
@@ -8531,7 +8528,9 @@
{ oid => '2509',
descr => 'deparse an encoded expression with pretty-print option',
proname => 'pg_get_expr', provolatile => 's', prorettype => 'text',
- proargtypes => 'pg_node_tree oid bool', prosrc => 'pg_get_expr_ext' },
+ proargtypes => 'pg_node_tree oid bool',
+ proargnames => '{expr,relation,pretty}', proargdefaults => '{false}',
+ prosrc => 'pg_get_expr' },
{ oid => '2510', descr => 'get the prepared statements for this session',
proname => 'pg_prepared_statement', prorows => '1000', proretset => 't',
provolatile => 's', proparallel => 'r', prorettype => 'record',
--
2.43.0
--eEEy3EHc57lA8spa
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
filename="v4-0006-Handle-pg_get_triggerdef-default-args-in-system_f.patch"
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
* [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix.
@ 2026-05-12 10:27 Antonin Houska <ah@cybertec.at>
0 siblings, 0 replies; 464+ messages in thread
From: Antonin Houska @ 2026-05-12 10:27 UTC (permalink / raw)
The fix: https://www.postgresql.org/message-id/77611.1778055944%40localhost
---
src/backend/access/transam/xact.c | 2 +
src/backend/commands/repack_worker.c | 2 +
src/test/isolation/isolationtester.c | 9 +-
.../expected/repack_running_xacts.out | 81 ++++++++++++
.../specs/repack_running_xacts.spec | 119 ++++++++++++++++++
5 files changed, 212 insertions(+), 1 deletion(-)
create mode 100644 src/test/modules/injection_points/expected/repack_running_xacts.out
create mode 100644 src/test/modules/injection_points/specs/repack_running_xacts.spec
diff --git a/src/backend/access/transam/xact.c b/src/backend/access/transam/xact.c
index 5586fbe5b07..b63ee166028 100644
--- a/src/backend/access/transam/xact.c
+++ b/src/backend/access/transam/xact.c
@@ -65,6 +65,7 @@
#include "utils/builtins.h"
#include "utils/combocid.h"
#include "utils/guc.h"
+#include "utils/injection_point.h"
#include "utils/inval.h"
#include "utils/memutils.h"
#include "utils/relmapper.h"
@@ -2428,6 +2429,7 @@ CommitTransaction(void)
* must be done _before_ releasing locks we hold and _after_
* RecordTransactionCommit.
*/
+ INJECTION_POINT("before-end-transaction", NULL);
ProcArrayEndTransaction(MyProc, latestXid);
/*
diff --git a/src/backend/commands/repack_worker.c b/src/backend/commands/repack_worker.c
index c40f8c98e06..6835626d677 100644
--- a/src/backend/commands/repack_worker.c
+++ b/src/backend/commands/repack_worker.c
@@ -26,6 +26,7 @@
#include "storage/ipc.h"
#include "storage/proc.h"
#include "tcop/tcopprot.h"
+#include "utils/injection_point.h"
#include "utils/memutils.h"
#define REPL_PLUGIN_NAME "pgrepack"
@@ -233,6 +234,7 @@ repack_setup_logical_decoding(Oid relid)
* Neither prepare_write nor do_write callback nor update_progress is
* useful for us.
*/
+ INJECTION_POINT("before-create-decoding-context", NULL);
ctx = CreateInitDecodingContext(REPL_PLUGIN_NAME,
NIL,
true,
diff --git a/src/test/isolation/isolationtester.c b/src/test/isolation/isolationtester.c
index 440c875b8ac..8f17ee412c9 100644
--- a/src/test/isolation/isolationtester.c
+++ b/src/test/isolation/isolationtester.c
@@ -216,15 +216,22 @@ main(int argc, char **argv)
* exactly expect concurrent use of test tables. However, autovacuum will
* occasionally take AccessExclusiveLock to truncate a table, and we must
* ignore that transient wait.
+ *
+ * If the session's backend is blocked, and if its background worker is
+ * waiting on an injection point, we assume that the injection point is
+ * the reason for the backend to be blocked. That's what we check in the
+ * second query of the UNION. XXX Should we use a separate query for that?
*/
initPQExpBuffer(&wait_query);
appendPQExpBufferStr(&wait_query,
+ "WITH blocking(res) AS ("
"SELECT pg_catalog.pg_isolation_test_session_is_blocked($1, '{");
/* The spec syntax requires at least one session; assume that here. */
appendPQExpBufferStr(&wait_query, conns[1].backend_pid_str);
for (i = 2; i < nconns; i++)
appendPQExpBuffer(&wait_query, ",%s", conns[i].backend_pid_str);
- appendPQExpBufferStr(&wait_query, "}')");
+ appendPQExpBufferStr(&wait_query, "}') UNION "
+ "SELECT pg_catalog.pg_isolation_test_session_is_blocked(pid, '{}') FROM pg_stat_activity WHERE leader_pid=$1) SELECT bool_or(res) FROM blocking");
res = PQprepare(conns[0].conn, PREP_WAITING, wait_query.data, 0, NULL);
if (PQresultStatus(res) != PGRES_COMMAND_OK)
diff --git a/src/test/modules/injection_points/expected/repack_running_xacts.out b/src/test/modules/injection_points/expected/repack_running_xacts.out
new file mode 100644
index 00000000000..271fe2b97cb
--- /dev/null
+++ b/src/test/modules/injection_points/expected/repack_running_xacts.out
@@ -0,0 +1,81 @@
+Parsed test spec with 5 sessions
+
+starting permutation: repack s3_assign_xid wakeup_bcdc s4_changes s4_attach s4_commit s3_commit s5_assign_xid wakeup_bet s5_commit check
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step repack:
+ REPACK (CONCURRENTLY) repack_test;
+ <waiting ...>
+step s3_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+
+step wakeup_bcdc:
+ SELECT injection_points_wakeup('before-create-decoding-context');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step s4_changes:
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+
+step s4_attach:
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+
+injection_points_set_local
+--------------------------
+
+(1 row)
+
+injection_points_attach
+-----------------------
+
+(1 row)
+
+step s4_commit:
+ COMMIT;
+ <waiting ...>
+step s3_commit:
+ COMMIT;
+
+step s5_assign_xid:
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+
+step wakeup_bet:
+ SELECT injection_points_wakeup('before-end-transaction');
+
+injection_points_wakeup
+-----------------------
+
+(1 row)
+
+step repack: <... completed>
+step s4_commit: <... completed>
+step s5_commit:
+ COMMIT;
+
+step check:
+ TABLE repack_test;
+
+i|j
+-+-
+(0 rows)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
+injection_points_detach
+-----------------------
+
+(1 row)
+
diff --git a/src/test/modules/injection_points/specs/repack_running_xacts.spec b/src/test/modules/injection_points/specs/repack_running_xacts.spec
new file mode 100644
index 00000000000..1f878514046
--- /dev/null
+++ b/src/test/modules/injection_points/specs/repack_running_xacts.spec
@@ -0,0 +1,119 @@
+setup
+{
+ CREATE EXTENSION injection_points;
+ CREATE TABLE repack_test(i int PRIMARY KEY, j int);
+ CREATE TABLE aux(i int);
+}
+
+teardown
+{
+ DROP TABLE repack_test;
+ DROP TABLE aux;
+ DROP EXTENSION injection_points;
+}
+
+session s1
+setup
+{
+ SELECT injection_points_attach('before-create-decoding-context', 'wait');
+}
+step repack
+{
+ REPACK (CONCURRENTLY) repack_test;
+}
+step check
+{
+ TABLE repack_test;
+}
+teardown
+{
+ SELECT injection_points_detach('before-create-decoding-context');
+}
+
+session s2
+step wakeup_bcdc
+{
+ SELECT injection_points_wakeup('before-create-decoding-context');
+}
+step wakeup_bet
+{
+ SELECT injection_points_wakeup('before-end-transaction');
+}
+
+session s3
+step s3_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (1);
+}
+step s3_commit
+{
+ COMMIT;
+}
+
+session s4
+step s4_changes
+{
+ BEGIN;
+ INSERT INTO repack_test(i, j) VALUES (1, 1);
+}
+# Do not attach in the setup section, that would be too soon.
+step s4_attach
+{
+ SELECT injection_points_set_local();
+ SELECT injection_points_attach('before-end-transaction', 'wait');
+}
+step s4_commit
+{
+ COMMIT;
+}
+teardown
+{
+ SELECT injection_points_detach('before-end-transaction');
+}
+
+session s5
+step s5_assign_xid
+{
+ BEGIN;
+ INSERT INTO aux VALUES (2);
+}
+step s5_commit
+{
+ COMMIT;
+}
+
+permutation
+repack
+# Assign XID so that a running transaction prevents the snapshot builder from
+# reaching CONSISTENT state immediately. It will wait for this to complete
+# after having reached BUILDING_SNAPSHOT.
+s3_assign_xid
+# Let the decoding setup start.
+wakeup_bcdc
+# Likewise, the snapshot builder will wait for the s4's xact to complete after
+# having reached FULL_SNAPSHOT. This is the problematic transaction, so let it
+# do some changes.
+s4_changes
+# Attach to the 'before-end-transaction' injection point that s4 will need
+# during commit.
+s4_attach
+# Only write commit record for s4, but do not remove the xact from procarray
+# yet. Thus the snapshot builder still needs to wait.
+s4_commit
+# Let the snapshot builder proceed to FULL_SNAPSHOT.
+s3_commit
+# Start another transaction so that CONSISTENT is not reached "directly",
+# i.e. due to no running transaction. It's important here that builder->xmin
+# does not advance.
+s5_assign_xid
+# Remove s4 xact from procarray, and thus reach the CONSISTENT state. Since
+# the COMMIT appeared in WAL too early (i.e. when the snapshot builder state
+# did not allow decoding of COMMIT records yet), the snapshot builder will
+# consider s4 running. This is also due to returning from
+# SnapBuildProcessRunningXacts() too early, w/o advancing builder->xmin.
+wakeup_bet
+# s5 is not needed anymore
+s5_commit
+# Show that the data changes performed by s4 are lost.
+check
--
2.47.3
--=-=-=--
^ permalink raw reply [nested|flat] 464+ messages in thread
end of thread, other threads:[~2026-05-12 10:27 UTC | newest]
Thread overview: 464+ messages (download: mbox mbox.gz follow: Atom feed)
-- links below jump to the message on this page --
2009-06-02 03:30 Re: explain analyze rows=%.0f Ron Mayer <rm_pg@cheapcomplexdevices.com>
2009-06-02 13:41 ` Simon Riggs <simon@2ndQuadrant.com>
2009-06-02 14:06 ` Robert Haas <robertmhaas@gmail.com>
2009-06-02 14:38 ` Tom Lane <tgl@sss.pgh.pa.us>
2009-06-02 14:56 ` Robert Haas <robertmhaas@gmail.com>
2020-03-21 01:47 [PATCH v1 1/2] Report wait event for cost-based vacuum delay Justin Pryzby <pryzbyj@telsasoft.com>
2025-12-09 19:17 [PATCH v2 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v4 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v3 6/7] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v3 6/7] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v8 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v8.1 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v8 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v4 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v5 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v2 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v7 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v1 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v7 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v1 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v8.1 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2025-12-09 19:17 [PATCH v5 5/6] Handle pg_get_expr default args in system_functions.sql Mark Wong <markwkm@gmail.com>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
2026-05-12 10:27 [PATCH] Test to demonstrate bug in commit 0d3dba38c7 and to verify a fix. Antonin Houska <ah@cybertec.at>
This inbox is served by agora; see mirroring instructions
for how to clone and mirror all data and code used for this inbox